* feat(telegram): stream LLM responses in real-time via sendMessageDraft
Implements real-time token streaming to Telegram using the sendMessageDraft
API (telego v1.6.0). Instead of showing only a "Thinking..." placeholder
until the full response arrives, users now see partial LLM output appear
in the chat as it's generated.
The streaming pipeline threads through all layers:
- StreamingProvider interface (providers/types.go): opt-in ChatStream()
method that receives an onChunk callback with accumulated text
- OpenAI-compatible SSE streaming (openai_compat/provider.go): parses
SSE events with stream:true, handles text deltas and tool call assembly
- Anthropic native streaming (anthropic/provider.go): uses SDK's
NewStreaming() for direct Anthropic API connections
- HTTPProvider delegation (http_provider.go): delegates ChatStream to
the underlying openai_compat provider
- StreamingCapable + Streamer interfaces (channels/interfaces.go):
opt-in channel capability like TypingCapable/PlaceholderCapable
- Telegram streamer (telegram/telegram.go): BeginStream returns a
telegramStreamer that throttles sendMessageDraft calls (3s/200 chars)
with graceful degradation on API errors
- StreamDelegate bridge (bus/bus.go): decouples agent loop from channel
manager without tight imports
- Manager integration (manager.go): implements StreamDelegate, tracks
streamActive state, coordinates with placeholder editing
- Agent loop (loop.go): uses ChatStream when both provider and channel
support streaming, cancels stream on tool calls, skips PublishOutbound
when Finalize already delivered the message
Graceful degradation:
- Bots without forum/topics mode: first sendMessageDraft error sets
failed=true, subsequent Updates become no-ops, Finalize still delivers
via SendMessage. User sees normal non-streaming behavior.
- Non-streaming providers: type assertion fails, falls back to Chat()
- Config opt-out: streaming.enabled (default true) in telegram config
Closes#1098
* fix(telegram): delete placeholder message when streaming delivers response
When streaming was active, the "Thinking..." placeholder message stayed
in the chat because preSend only deleted the tracking entry without
removing the actual Telegram message. Now preSend deletes the placeholder
via the new MessageDeleter interface when streamActive is set.
* refactor(streaming): remove dead code and simplify streaming wiring
- Delete unused Anthropic ChatStream/parseStream (-131 lines) — factory
creates HTTPProvider for all OpenAI-compat providers including OpenRouter
- Simplify runLLMIteration from 4 to 3 return values (remove unused
streamed bool)
- Replace managerStreamer struct with finalizeHookStreamer using embedding
(Update/Cancel promoted, only Finalize overridden)
* fix(streaming): skip streamer acquisition when SendResponse is false
Heartbeat messages set SendResponse=false but the streaming path
was unconditionally acquiring a streamer, causing HEARTBEAT_OK to
leak to Telegram via streamer.Finalize().
* fix(streaming): guard streamer for non-sendable messages, add streaming config
Skip streamer acquisition for heartbeat (NoHistory=true), preventing
HEARTBEAT_OK from leaking to Telegram via streamer.Finalize().
Add streaming.enabled to Telegram defaults and example config.
* feat(telegram): stream LLM responses in real-time via sendMessageDraft
Implements real-time token streaming to Telegram using the sendMessageDraft
API (telego v1.6.0). Instead of showing only a "Thinking..." placeholder
until the full response arrives, users now see partial LLM output appear
in the chat as it's generated.
The streaming pipeline threads through all layers:
- StreamingProvider interface (providers/types.go): opt-in ChatStream()
method that receives an onChunk callback with accumulated text
- OpenAI-compatible SSE streaming (openai_compat/provider.go): parses
SSE events with stream:true, handles text deltas and tool call assembly
- Anthropic native streaming (anthropic/provider.go): uses SDK's
NewStreaming() for direct Anthropic API connections
- HTTPProvider delegation (http_provider.go): delegates ChatStream to
the underlying openai_compat provider
- StreamingCapable + Streamer interfaces (channels/interfaces.go):
opt-in channel capability like TypingCapable/PlaceholderCapable
- Telegram streamer (telegram/telegram.go): BeginStream returns a
telegramStreamer that throttles sendMessageDraft calls (3s/200 chars)
with graceful degradation on API errors
- StreamDelegate bridge (bus/bus.go): decouples agent loop from channel
manager without tight imports
- Manager integration (manager.go): implements StreamDelegate, tracks
streamActive state, coordinates with placeholder editing
- Agent loop (loop.go): uses ChatStream when both provider and channel
support streaming, cancels stream on tool calls, skips PublishOutbound
when Finalize already delivered the message
Graceful degradation:
- Bots without forum/topics mode: first sendMessageDraft error sets
failed=true, subsequent Updates become no-ops, Finalize still delivers
via SendMessage. User sees normal non-streaming behavior.
- Non-streaming providers: type assertion fails, falls back to Chat()
- Config opt-out: streaming.enabled (default true) in telegram config
Closes#1098
* fix(telegram): delete placeholder message when streaming delivers response
When streaming was active, the "Thinking..." placeholder message stayed
in the chat because preSend only deleted the tracking entry without
removing the actual Telegram message. Now preSend deletes the placeholder
via the new MessageDeleter interface when streamActive is set.
* refactor(streaming): remove dead code and simplify streaming wiring
- Delete unused Anthropic ChatStream/parseStream (-131 lines) — factory
creates HTTPProvider for all OpenAI-compat providers including OpenRouter
- Simplify runLLMIteration from 4 to 3 return values (remove unused
streamed bool)
- Replace managerStreamer struct with finalizeHookStreamer using embedding
(Update/Cancel promoted, only Finalize overridden)
* fix(streaming): skip streamer acquisition when SendResponse is false
Heartbeat messages set SendResponse=false but the streaming path
was unconditionally acquiring a streamer, causing HEARTBEAT_OK to
leak to Telegram via streamer.Finalize().
* fix(streaming): guard streamer for non-sendable messages, add streaming config
Skip streamer acquisition for heartbeat (NoHistory=true), preventing
HEARTBEAT_OK from leaking to Telegram via streamer.Finalize().
Add streaming.enabled to Telegram defaults and example config.
* fix(picoclaw): add missing closing brace for StreamingProvider interface
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix: resolve golangci-lint formatting issues
Fix gci import ordering in telegram and anthropic provider, and break
long function signature in openai_compat provider to satisfy golines.
* fix: address code review feedback on streaming PR
- Deduplicate Streamer interface: alias channels.Streamer to bus.Streamer
to prevent type drift across packages
- Increase SSE scanner buffer to 10MB max to handle large single-line
responses that exceed bufio.Scanner's 64KB default
- Switch draftID generation from math/rand to crypto/rand for
collision-resistant random IDs
- Add context cancellation check in SSE parsing loop so cancelled
streams stop processing immediately
- Log Finalize failures with chat_id and content length for debugging
silent message delivery failures
* feat: make streaming throttle interval and min growth configurable
Move hardcoded streamThrottleInterval (3s) and streamMinGrowth (200)
into StreamingConfig so they can be tuned per deployment via config
or environment variables.
* fix(telegram): use parseTelegramChatID in DeleteMessage and BeginStream
These two functions called undefined parseChatID. Use
parseTelegramChatID with _ for the unused threadID instead of adding
a wrapper function. Fixes all three CI checks.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(streaming): set streamActive only after successful Finalize
Move onFinalize hook to run after Streamer.Finalize succeeds, so that
if Finalize fails the streamActive flag stays false and the regular
placeholder fallback path remains available.
Addresses review feedback from @alexhoshina.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* feat(pico): add pico_client outbound WebSocket channel
Add a client-mode counterpart to the existing pico server channel.
pico_client connects to a remote Pico Protocol WebSocket server,
enabling picoclaw to bridge messages with external Pico-compatible
services.
Includes config, factory registration, manager wiring, 8 unit tests,
and a minimal echo-server example for interactive testing.
* fix(pico): address PR #1198 review — goroutine leak, race, auth
- Add per-connection context cancel to picoConn to prevent pingLoop
goroutine leak on disconnect
- Re-acquire mutex in StartTyping stop closure to avoid stale conn race
- Remove query-param token auth from echo server (header-only)
- Move ListenAndServe to main goroutine where log.Fatal is safe
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix: replace ConsumeInbound with InboundChan select in client test
MessageBus does not expose a ConsumeInbound method. Use a select on
InboundChan() with context cancellation, matching the pattern used in
the bus package tests.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Refactor agent loop execution around runTurn, add explicit turn state and interrupt semantics, and automatically continue queued steering that misses the current turn boundary.
* feat(feishu): add interactive card message parsing
Add support for parsing inbound Feishu interactive card messages.
When a user sends a card message, the text content is now extracted
and passed to the LLM for processing.
- Add extractCardText() to recursively extract text from card JSON
- Support both JSON 1.0 (legacy) and JSON 2.0 schema formats
- Handle nested elements: header, body, actions, columns
- Extract text from markdown, lark_md, and plain_text elements
- Add comprehensive unit tests for card parsing
Fixes #<issue_number>
💘 Generated with Crush
Assisted-by: GLM-5 via Crush <crush@charm.land>
* feat(feishu): extract and download images from interactive cards
When receiving interactive card messages, extract embedded images
(img_key, src, icon_key) and download them for LLM processing.
- Add extractCardImageKeys() to recursively extract image keys from card JSON
- Support img elements (img_key, src) and icon elements (icon_key)
- Update downloadInboundMedia() to handle MsgTypeInteractive
- Add comprehensive unit tests for image extraction
Images are downloaded and stored via MediaStore, then appended to
the message content as [image: photo] tags for LLM visibility.
💘 Generated with Crush
Assisted-by: GLM-5 via Crush <crush@charm.land>
* fix(feishu): simplify card parsing - pass raw JSON, only extract images
Address review feedback: text extraction cannot exhaustively handle all
card formats (i18n_elements, div.fields, etc.). Pass raw JSON to LLM
instead - same approach as MsgTypePost. Only image extraction remains
as images must be downloaded for LLM to process.
- Remove extractCardText() and helper functions
- extractContent() now returns raw JSON for MsgTypeInteractive
- Keep extractCardImageKeys() for downloading embedded images
- Update tests to expect raw JSON for interactive cards
* fix(feishu): don't append media tags to interactive card JSON
Appending media tags like "[attachment]" to raw JSON content produces
invalid JSON format. For interactive cards, the JSON already contains
image information and media refs are downloaded separately.
- Skip appendMediaTags for MsgTypeInteractive to preserve valid JSON
- Add test case for interactive card with images
* fix(feishu): filter out external URLs from card image extraction
Only Feishu-hosted image keys (img_xxx, icon_xxx) can be downloaded via
the Feishu API. External URLs in src field (https://...) should be
filtered out to avoid download failures.
- Add isFeishuImageKey() to detect Feishu-hosted keys vs external URLs
- Update extractImageKeysRecursive to skip external URLs in src field
- Add tests for external URL filtering and mixed scenarios
* feat(feishu): support downloading external images from interactive cards
Previously only Feishu-hosted images (img_key, icon_key) could be
downloaded. Now external URLs in src field are also downloaded via
HTTP and made available to the LLM.
- extractCardImageKeys now returns two slices: Feishu keys and external URLs
- Add downloadExternalImage to download images from HTTP URLs
- Update downloadInboundMedia to handle both Feishu API and HTTP downloads
- Update tests for new function signature
* fix(feishu): use HTTP client with timeout for external image downloads
Replaced http.DefaultClient with a client that has a 30-second timeout
to prevent hanging on unresponsive external URLs.
Generated with Crush
Assisted-by: GLM-5 via Crush <crush@charm.land>
* fix(feishu): resolve lint errors for shadow and formatting
- Rename err variables to avoid shadowing in downloadExternalImage
- Fix struct field alignment in TestExtractCardImageKeys
Generated with Crush
Assisted-by: GLM-5 via Crush <crush@charm.land>
* refactor(feishu): pass external image URLs to LLM instead of downloading
Instead of downloading external images from interactive cards, pass
the URLs directly to LLM. This reduces network overhead and lets
vision-capable models fetch images as needed.
- Remove downloadExternalImage function
- Append external URLs to card content for LLM processing
- Only download Feishu-hosted images via API
💘 Generated with Crush
Assisted-by: GLM-5 via Crush <crush@charm.land>
* fix(feishu): add blank line between functions for gci formatting
* fix(feishu): keep interactive card content as valid JSON
Changed newTestAgentLoop calls from using 3 blank identifiers to 2 by
assigning the unused provider parameter and explicitly marking it as
unused with `_ = provider`. This fixes the dogsled linter violations
that were causing CI failures.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
When both providers and model_list are configured, model_list entries
with empty api_key or api_base now automatically inherit from the
matching provider (matched by protocol prefix in the Model field).
Example: a model_list entry with model='deepseek/deepseek-chat' and
no api_key will inherit from providers.deepseek.api_key.
Explicit model_list values always take precedence.
Changes:
- Add InheritProviderCredentials() in migration.go
- Call it in LoadConfig() after provider-to-model-list conversion
- Add protocolProviderMapping for all 25 supported protocols
- 6 new tests covering inheritance, precedence, and edge cases
Closes#1635
* feat(wecom): add WebSocket long-connection support for WeCom AI Bot
- Introduced WeComAIBotWSChannel to handle WebSocket connections.
- Updated NewWeComAIBotChannel to prioritize WebSocket mode when BotID and Secret are provided.
- Enhanced WeComAIBotConfig to include BotID and Secret for WebSocket mode.
- Implemented message handling for text, image, voice, and mixed messages in WebSocket mode.
- Added tests for WebSocket mode functionality and ensured backward compatibility with webhook mode.
- Refactored existing code to improve clarity and maintainability.
* feat(wecom): implement periodic processing hints and enforce WeCom stream deadline
* feat(wecom): update WeCom AI Bot setup instructions and configuration parameters
* feat(wecom): enhance WeCom AI Bot with image handling and media support
* feat(wecom): refactor WeCom AI Bot task management to use req_id for concurrent message handling
* feat(wecom): refactor WeCom AI Bot to manage request states and late replies
* feat(wecom): add response timeout handling and improve WebSocket command acknowledgment
* fix(wecom): improve error handling for late reply proactive push delivery
* refactor(wecom): reorganize WeCom AI Bot configuration fields for improved readability
* fix(wecom): update error message for websocket delivery failure in late reply proactive push
* feat(wecom): implement shared HTTP clients for WeCom image handling and response URL posting
* refactor(wecom): simplify image download and storage process in storeWSImage
* fix(wecom): improve error logging for WebSocket message handling and proactive push delivery
* fix(wecom): enhance WebSocket connection stability and task cancellation handling
* fix(wecom): improve WS image message handling by ensuring proper error response and initializing mediaRefs
* feat(wecom): enhance WeCom AIBot WebSocket handling with message deduplication and support for file and video messages
* refactor(wecom): rename image handling functions to media handling and enhance media type support
* feat(wecom): implement byte-aware content splitting for WeCom AI Bot stream messages
* refactor(wecom): remove max message length constraint from WeCom AIBot WS channel
Add nil checks in NewSpawnTool and NewSubagentTool constructors to
handle nil manager gracefully. Fix spelling errors (cancelled->canceled)
and remove unused test code. Update tests to use mock spawner.
Replace hardcoded constants with config-driven parameters in agents.defaults:
- MaxDepth, MaxConcurrent, DefaultTimeout, DefaultTokenBudget, ConcurrencyTimeout
- Support JSON config and env vars (PICOCLAW_AGENTS_DEFAULTS_SUBTURN_*)
- Add getSubTurnConfig() for runtime config resolution with defaults
- Apply defaultTokenBudget when no explicit budget is provided
Rationale: SubTurn is agent execution infrastructure, not a tool, so it belongs
in agents.defaults rather than tools config.
Example:
{
"agents": {
"defaults": {
"subturn": {
"max_depth": 5,
"max_concurrent": 10,
"default_timeout_minutes": 10
}
}
}
}
Add ActualSystemPrompt and InitialMessages fields to SubTurnConfig to enable
stateful worker context passing across multiple evaluation iterations.
Changes:
- Add ActualSystemPrompt field to separate system role from user task description
- Add InitialMessages field to preload ephemeral session history before agent loop starts
- Add Messages field to ToolResult for carrying session history (internal use, not serialized)
- Update runTurn to inject system prompt and preload history from InitialMessages
- Update AgentLoopSpawner to map new fields from tools.SubTurnConfig to agent.SubTurnConfig
This enables the evaluator-optimizer execution strategy in team tool to maintain
worker context across iterations while keeping SubTurn isolation intact.
* feat(config): support multiple API keys for failover
Add api_keys field to ModelConfig to support multiple API keys with
automatic failover. When multiple keys are configured, they are expanded
into separate model entries with fallbacks set up for key-level failover.
Example config:
{
"model_name": "glm-4.7",
"model": "zhipu/glm-4.7",
"api_keys": ["key1", "key2", "key3"]
}
Expands internally to:
- glm-4.7 (key1) -> fallbacks: [glm-4.7__key_1, glm-4.7__key_2]
- glm-4.7__key_1 (key2)
- glm-4.7__key_2 (key3)
Backward compatible: single api_key still works as before.
* fix(providers): change cooldown tracking from provider to ModelKey
This enables proper key-switching when multiple API keys share the same
provider. Previously, when one key failed, all keys were blocked because
cooldown was tracked per-provider.
Now each (provider, model) combination has independent cooldown, allowing
fallback to alternate keys when one is rate limited.
Includes TestMultiKeyWithModelFallback and related failover tests.
* feat(feishu): add Lark (international) support via IsLark config field
Add IsLark field to FeishuConfig to switch between Feishu and Lark
domains. Also fix domain inconsistency where WS client defaulted to
LarkBaseUrl while HTTP client used FeishuBaseUrl.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* docs: update documentation and web UI for Lark support
Add is_lark field to config example, feishu docs, i18n translations,
and web frontend form.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* fix(tools): propagate tool registry to subagents via Clone
SubagentManager was created with an empty ToolRegistry and SetTools()
was never called, causing all subagent tool invocations to fail with
"tool not found". This was a regression from the multi-agent refactor.
Fix: clone the parent agent's tool registry into the subagent manager
after creation but before spawn/spawn_status registration — giving
subagents access to file, exec, web, and other tools while preventing
recursive subagent spawning.
- Add ToolRegistry.Clone() for independent shallow copies
- Call subagentManager.SetTools(agent.Tools.Clone()) in registerSharedTools
- Add tests for Clone isolation, empty clone, and hidden tool state
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(tools): fix cron_test build error and add TTL clone test
- Fix cron_test.go:229 — replace non-existent SubscribeOutbound(ctx)
with select on OutboundChan(), matching the MessageBus channel API
- Add TestToolRegistry_Clone_PreservesTTLValue per reviewer feedback
- Add version reset note to Clone() doc comment
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* feat(agent): add structured agent definition loader
Parse AGENT.md frontmatter into a runtime definition and pair it with SOUL.md while keeping a legacy AGENTS.md fallback for transition.
Refs #1218
* refactor(agent): build context from structured agent files
Use AGENT.md and SOUL.md as the structured bootstrap source, ignore IDENTITY.md for structured agents, remove USER.md from the new context flow, and update pkg/agent tests accordingly.
Refs #1218
* refactor(onboard): switch workspace templates to AGENT.md
Replace the legacy AGENTS.md, IDENTITY.md, and USER.md templates with a structured AGENT.md plus SOUL.md, and update the onboard template test to assert the new generated files.
Refs #1218
* docs(readme): update workspace layout for AGENT.md
Refresh the documented workspace tree across the README translations so onboarding now points to AGENT.md and SOUL.md instead of the retired AGENTS.md, IDENTITY.md, and USER.md files.
Refs #1218
* feat(agent): restore workspace USER.md context
* docs(readme): document workspace USER.md layout
* fix: sort agent definition imports for gci
When building parameters for Anthropic API calls, tool calls with empty
names would cause 400 Bad Request errors with the message:
'tool_use.name: String should have at least 1 character'
This fix adds a check to skip tool calls that have empty names, preventing
the API error and allowing the conversation to continue normally.
Fixes#1658
The Lark SDK v3's built-in token retry loop does not clear stale tokens
from cache when the server returns error 99991663 (tenant_access_token
invalid), causing all API calls to fail until the token naturally
expires (~2 hours).
- Add tokenCache struct (implementing larkcore.Cache) with
Get/Set/InvalidateAll methods and proper expired-entry cleanup
- Wire custom cache into lark.NewClient via WithTokenCache()
- Add invalidateTokenOnAuthError helper called in all API methods
* Add Novita provider support
- Add 'novita' prefix to normalizeModel switch in openai_compat provider
- Add Novita provider to all_supported_vendors table in README.md
- Add test cases for Novita model prefix stripping
Novita endpoint: https://api.novita.ai/openai
Default models: deepseek/deepseek-v3.2, zai-org/glm-5, minimax/minimax-m2.5
* feat: complete Novita provider integration
* chore: drop README changes from Novita PR
* fix: remove duplicate function declarations in openai_compat provider
The functions buildToolsList, SupportsNativeSearch, and isNativeSearchHost
were declared twice, causing compilation failures in all CI checks.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix: break long line in novita test to satisfy golines linter
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
- Critical flag was declared but never acted on; non-critical SubTurns
now break out of the iteration loop when IsParentEnded() returns true
- tools.SubTurnConfig was missing Critical/Timeout/MaxContextRunes,
making those fields unreachable from the tools layer; added fields and
wired them through AgentLoopSpawner.SpawnSubTurn
- Removed subTurnResults sync.Map from AgentLoop — it was a redundant
alias for the same channel already stored in turnState.pendingResults;
dequeuePendingSubTurnResults now reads directly via activeTurnStates
- Replace hardcoded concurrencySem size 5 with maxConcurrentSubTurns constant
- Update affected tests to match new dequeuePendingSubTurnResults API
* fix(telegram): improve HTML chunking and preserve word boundaries
* fix(telegram): address copilot feedback, filter empty chunks and add word-boundary regression test
* style(telegram): fix gofmt and gci lint errors in tests
* fix to feedback
- Added `/subagents` platform command to visualize the active task tree.
- Implemented GetAllActiveTurns and FormatTree in AgentLoop to support cross-session observability.
- Fixed a bug where sub-turns spawned via tools were not registered in the global `activeTurnStates` map, making them invisible to system queries.
- Enhanced tree rendering logic to identify and display "orphaned" subagents (children that outlive their parent turns).
- Registered the new command in `builtin.go` and injected the turn state provider into the commands runtime.
Modified Files:
- pkg/agent/turn_state.go: Added TurnInfo snapshotting and recursive tree formatting.
- pkg/agent/loop.go: Injected GetActiveTurn hook and implemented multi-root forest rendering.
- pkg/agent/subturn.go: Added child turn registration into activeTurnStates.
- pkg/commands/cmd_subagents.go: New command implementation.
- pkg/commands/builtin.go: Command registration.
This commit addresses several critical concurrency and state management bugs within the SubTurn execution and delivery logic.
1. Fix Goroutine Leak & Deadlock in deliverSubTurnResult:
- Replaced non-blocking select with a safe blocking select that listens to `resultChan` and a new `<-parentTS.Finished()` channel.
- This ensures results are not arbitrarily dropped when the channel is full (preventing orphaned valid results), while also guaranteeing the child goroutine safely unblocks and exits if the parent finishes execution early.
2. Prevent "Send on Closed Channel" Fatal Panics:
- Removed `close(pendingResults)` and `drainPendingResults` from `turnState.Finish()`.
- The pendingResults channel is now naturally garbage collected, completely eliminating the race condition panic when a child attempts delivery at the exact moment the parent finishes.
- Added a `defer recover()` failsafe inside deliverSubTurnResult to gracefully emit Orphan events in extreme edge cases.
3. Fix Truncation Recovery Prompt Drop:
- Fixed the runTurn truncation retry logic by introducing an explicit `promptAlreadyAdded` boolean.
- Ensures that the dynamically generated `recoveryPrompt` is correctly injected into the LLM history sequence on subsequent iterations, adhering to API roles without duplicating arrays.
4. Test Suite Stabilization:
- Fixed TestDeliverSubTurnResultNoDeadlock to accurately wait for deterministic deliveries instead of racing timeouts.
- Replaced defunct closed-channel tests with TestFinishedChannelClosedState matching the new Finished() mechanism.
- Fixed the Finish(true) parameter in TestGrandchildAbort_CascadingCancellation to correctly validate Context cascade behavior.
- All tests now pass cleanly without hanging or emitting false positives.
Problem:
During subturn context limit or truncation recoveries, the recovery loops repeatedly
called `runAgentLoop` with the same or modified `UserMessage`. Because `runAgentLoop`
unconditionally adds the `UserMessage` to the session history, this resulted in:
1. Duplicate User Messages polluting the history upon `context_length_exceeded` retries.
2. The possibility of injecting empty User Messages if `opts.UserMessage` was artificially blanked out to work around the duplication.
3. Messy or duplicate entries during `finish_reason="truncated"` recovery injections.
Solution:
- Introduce `SkipAddUserMessage` boolean to `processOptions` to explicitly control whether the agent loop should write the user prompt to history.
- Add an explicit `opts.UserMessage != ""` check in `runAgentLoop` to prevent polluting history with empty message content.
- In `subturn.go`'s recovery loop, set `SkipAddUserMessage: contextRetryCount > 0` to skip writing the user message on context
* config: add prefer_native and NativeSearchCapable for model-native search
* providers: implement native web search for OpenAI and Codex
* agent: use provider-native search when prefer_native and supported
* tests: add coverage for model-native search
* fix: Golang lint errors
* fix: update the code based on the review
* fix: update codex_provider_test
* fix: Fixed the bug where the bus was closed and consumers had unfinished messages.
* fix: remove unnecessary blank line in Close method
* fix: refactor message bus and channel handling for improved performance and reliability
* fix: improve message handling and bus closure logic for better reliability
* fix: reduce sleep duration in agent loop for improved responsiveness
* fix the test case
* feat(cron): enhance CronService with wake channel and improve job scheduling logic
* fix(cron): update file permission mode to use octal notation in test and fix some lint errors
* fix(cron): improve wake channel handling and enhance concurrency in tests
Problem:
When parent turn finishes early, all child SubTurns receive "context canceled"
error,because child context was derived from parent context.
Solution:
Implement a lifecycle management system that distinguishes between:
- Graceful finish (Finish(false)): signals parentEnded, children continue
- Hard abort (Finish(true)): immediately cancels all children
Changes:
- turn_state.go:
- Add parentEnded atomic.Bool to signal parent completion
- Add parentTurnState reference for IsParentEnded() checks
- Modify Finish(isHardAbort bool) to distinguish abort types
- subturn.go:
- Add Critical bool to SubTurnConfig (Critical SubTurns continue after parent ends)
- Add Timeout time.Duration for SubTurn self-protection
- Use independent context (context.Background()) instead of derived context
- SubTurns check IsParentEnded() to decide whether to continue or exit
- loop.go:
- Call Finish(false) for normal completion (graceful)
- Add IsParentEnded() check in LLM iteration loop
- steering.go:
- HardAbort calls Finish(true) to immediately cancel children
Behavior:
- Normal finish: parentEnded=true, children continue, orphan results delivered
- Hard abort: all children cancelled immediately via context
- Critical SubTurns: continue running after parent finishes gracefully
- Non-Critical SubTurns: can exit gracefully when IsParentEnded() returns true
Includes JSONL session persistence (#1170), spawn_status tool, Azure provider,
credential encryption, and various fixes. SubTurn features preserved and
integrated with new spawn_status functionality.
- add a dedicated exec settings section in the config page
- support timeout and custom allow/deny regex patterns for exec
- validate custom exec regex patterns in the config API
- block cron command scheduling and execution when exec is disabled
- update tests and i18n strings for the new command settings
* feat(gateway): support hot reload and empty startup
- extract gateway runtime into pkg/gateway
- add gateway.hot_reload config with default and example values
- allow starting the gateway without a default model via --allow-empty
- stop treating missing enabled channels as a startup error
- update related tests
* feat: replace gateway SSE updates with polling-based state sync
- remove gateway SSE broadcasting and event endpoint
- add polling-based gateway status refresh with stopping state handling
- detect when gateway restart is required after default model changes
- resolve gateway health and websocket proxy targets from configured host
- update gateway UI labels and add backend/frontend test coverage
* feat(tools): add SpawnStatusTool for reporting subagent statuses
* feat(tools): enhance SpawnStatusTool to restrict task visibility by conversation context
* feat(tests): add Unicode result truncation and channel filtering tests for SpawnStatusTool
* Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
* feat(tools): enhance SpawnStatusTool with task ID validation and sorting by creation timestamp
* feat(tools): update SpawnStatusTool description and parameter documentation for clarity
* refactor(tests): improve comments for clarity in ChannelFiltering test case
* fix(tools): update no subagents message for clarity and remove unnecessary locking in runTask
* fix(tools): improve description clarity for SpawnStatusTool regarding task context
* feat(tools): add spawn_status tool configuration and registration
* Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
* fix(agent): improve subagent management for spawn and spawn_status tools
* Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
* Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
* fix(tests): update ResultTruncation_Unicode test to use valid CJK character
---------
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
Co-authored-by: lxowalle <83055338+lxowalle@users.noreply.github.com>
- add defensive nil check for tool call Arguments field
- replace nil input with empty object to comply with Anthropic spec
- prevent API errors when GLM models return null input in tool_use blocks
Zhipu AI's GLM series models may return tool_use blocks with null input field,
which causes their API to reject subsequent requests with error:
"ClaudeContentBlockToolResult object has no attribute id"
This fix ensures compatibility by converting nil inputs to empty objects {},
matching the Anthropic Messages API specification while maintaining backward
compatibility with other providers.
When the entire session history is a single Turn (e.g. one user message
followed by a massive tool response), findSafeBoundary returns 0 and
forceCompression previously did nothing — leaving the agent stuck in
a context-exceeded retry loop.
Now falls back to keeping only the most recent user message when no
safe Turn boundary exists. This breaks Turn atomicity as a last resort
but guarantees the agent can recover.
Also updates docs/agent-refactor/context.md to document this behavior.
Ref #1490
- add tools.cron.allow_command config with a default value of true
- require command_confirm only when cron command execution is disabled
- expose cron command permission and timeout settings in the config UI
- add backend tests and update i18n strings
- Fix synchronous SubTurn calls placing results in pendingResults channel,
causing double delivery. Now only async calls (Async=true) use the channel.
- Move deliverSubTurnResult into defer to ensure result delivery even when
runTurn panics. Add TestSpawnSubTurn_PanicRecovery to verify.
- Fix ContextWindow incorrectly set to MaxTokens; now inherits from
parentAgent.ContextWindow.
- Add TestSpawnSubTurn_ResultDeliverySync to verify sync behavior.
Critical fixes (5):
- Fix turnState hierarchy corruption in nested SubTurns by checking context
before creating new root turnState in runAgentLoop
- Fix deadlock risk in deliverSubTurnResult by separating lock and channel ops
- Fix session rollback race in HardAbort by calling Finish() before rollback
- Fix resource leak by closing pendingResults channel in Finish() with recovery
- Add thread-safety docs for childTurnIDs and isFinished fields
Medium priority fixes (5):
- Move globalTurnCounter to AgentLoop.subTurnCounter to prevent ID conflicts
- Improve semaphore acquisition to ensure release even on early validation failures
- Document design choice: ephemeral sessions start empty for complete isolation
- Add final poll before Finish() to capture late-arriving SubTurn results
- Remove duplicate channel registration in spawnSubTurn to fix timing issues
Testing:
- Add 6 new tests covering hierarchy, deadlock, ordering, channel lifecycle,
final poll, and semaphore behavior
- All 12 SubTurn tests passing with race detector
This resolves 10 critical and medium issues (5 race conditions, 2 resource leaks,
3 timing issues) identified in code review, bringing SubTurn to production-ready state.
- Fix turnState hierarchy corruption when SubTurns recursively call runAgentLoop
by checking context for existing turnState before creating new root
- Fix deadlock risk in deliverSubTurnResult by separating lock and channel operations
- Fix session rollback race in HardAbort by calling Finish() before rollback
- Fix resource leak by closing pendingResults channel in Finish() with panic recovery
- Add thread-safety documentation for childTurnIDs and isFinished fields
- Move globalTurnCounter to AgentLoop.subTurnCounter to prevent ID conflicts
- Improve semaphore acquisition to ensure release even on early validation failures
- Document design choice: ephemeral sessions start empty for complete isolation
- Add 5 new tests: hierarchy, deadlock, order, channel close, and semaphore
- Add initialHistoryLength field to turnState to snapshot session state at turn start
- Save initial history length in runAgentLoop when creating root turnState
- Implement session rollback in HardAbort via SetHistory, truncating to initial length
- Add TestHardAbortSessionRollback to verify history rollback after abort
- Import providers package in subturn_test.go for Message type
This ensures that when a user triggers hard abort, all messages added during
the aborted turn are discarded, restoring the session to its pre-turn state.
- Add maxConcurrentSubTurns constant (5) and concurrencySem channel to turnState
- Acquire/release semaphore in spawnSubTurn to limit concurrent child turns per parent
- Add activeTurnStates sync.Map to AgentLoop for tracking root turn states by session
- Implement HardAbort(sessionKey) method to trigger cascading cancellation via turnState.Finish()
- Register/unregister root turnState in runAgentLoop for hard abort lookup
- Add TestSubTurnConcurrencySemaphore to verify semaphore capacity enforcement
- Add TestHardAbortCascading to verify context cancellation propagates to child turns
- Add subTurnResults sync.Map to AgentLoop for per-session channel tracking
- Add register/unregister/dequeue methods in steering.go
- Poll SubTurn results in runLLMIteration at loop start and after each tool,
injecting results as [SubTurn Result] messages into parent conversation
- Initialize root turnState in runAgentLoop, propagate via context
(withTurnState/turnStateFromContext), call rootTS.Finish() on completion
- Wire Spawn Tool to spawnSubTurn via SetSpawner in registerSharedTools,
recovering parentTS from context for proper turn hierarchy
- Refactor subagent.go to use SetSpawner pattern
- Add TestSubTurnResultChannelRegistration and TestDequeuePendingSubTurnResults
Session history only stores user/assistant/tool messages — the system
prompt is built dynamically by BuildMessages. Remove the incorrect
system message from TestAgentLoop_ContextExhaustionRetry test data
to match the real data model that forceCompression operates on.
When the entire history is a single Turn (one user message followed by
tool calls and responses, no subsequent user message), the only Turn
boundary is at index 0. Previously the fallback returned targetIndex,
which could land on a tool or assistant message — splitting the Turn.
Return 0 instead, so callers (forceCompression, summarizeSession) see
mid <= 0 and skip compression rather than cutting inside the Turn.
Two estimation bugs fixed:
1. Media tokens were added to the chars accumulator before the chars*2/5
conversion, resulting in 256*2/5=102 tokens per item instead of 256.
Fix: add media tokens directly to the final token count, bypassing
the character-based heuristic.
2. estimateMessageTokens counted both tc.Name and tc.Function.Name for
tool calls, but providers only send one (OpenAI-compat uses
function.name, Anthropic uses tc.Name). Fix: count tc.Function.Name
when Function is present, fall back to tc.Name only otherwise.
Also fix i18n hint text: "auto-detect" was misleading — the backend
uses a 4x max_tokens heuristic, not actual model detection.
Introduce parseTurnBoundaries() which identifies each Turn start index
in the session history. A Turn is a complete "user input → LLM iterations
→ final response" cycle (as defined in the agent refactor design #1316).
findSafeBoundary now uses Turn boundaries instead of raw role-scanning,
making the intent explicit: "find the nearest Turn boundary."
forceCompression drops the oldest half of Turns (not arbitrary messages),
which is simpler and more intuitive. The Turn-based approach naturally
prevents splitting tool-call sequences since each Turn is atomic.
Add tests that reflect actual session data shape: history starts with
user messages (no system prompt), includes chained tool-call sequences,
reasoning content, and media items. Exercises the proactive budget check
path with BuildMessages-style assembled messages.
estimateMessageTokens now counts ReasoningContent (extended thinking /
chain-of-thought) which can be substantial and is persisted in session
history. Media items get a fixed per-item overhead (256 tokens) since
actual cost depends on provider-specific image tokenization.
Session history (GetHistory) contains only user/assistant/tool messages.
The system prompt is built dynamically by BuildMessages and is never
stored in session. The previous code incorrectly treated history[0] as
a system prompt, skipping the first user message and appending a
compression note to it.
Fix: operate on the full history slice, and record the compression
note in the session summary (which BuildMessages already injects into
the system prompt) rather than modifying any history message.
Separate context_window from max_tokens — they serve different purposes
(input capacity vs output generation limit). The previous conflation caused
premature summarization or missed compression triggers.
Changes:
- Add context_window field to AgentDefaults config (default: 4x max_tokens)
- Extract boundary-safe truncation helpers (isSafeBoundary, findSafeBoundary)
into context_budget.go — pure functions with no AgentLoop dependency
- forceCompression: align split to safe boundary so tool-call sequences
(assistant+ToolCalls → tool results) are never torn apart
- summarizeSession: use findSafeBoundary instead of hardcoded keep-last-4
- estimateTokens: count ToolCalls arguments and ToolCallID metadata,
not just Content — fixes systematic undercounting in tool-heavy sessions
- Add proactive context budget check before LLM call in runAgentLoop,
preventing 400 context-length errors instead of reacting to them
- Add estimateToolDefsTokens for tool definition token cost
Closes#556, closes#665
Ref #1439
- Replace duplicate types (ToolResult/Session/Message) with real project types
- Implement ephemeralSessionStore satisfying session.SessionStore interface
- Connect runTurn to real AgentLoop via runAgentLoop + AgentInstance
- Fix subturn_test.go to match updated signatures and types
Co-Authored-By: Claude Sonnet 4 <noreply@anthropic.com>
* feat(credential): add AES-GCM encryption, SecureStore, and onboard keygen
- pkg/credential: new package with AES-256-GCM enc:// credential format,
HKDF-SHA256 key derivation (passphrase + optional SSH key binding),
ErrPassphraseRequired / ErrDecryptionFailed sentinel errors,
and PassphraseProvider hook for runtime passphrase injection
- pkg/credential/store: lock-free SecureStore via atomic.Pointer[string];
passphrase never written to disk or os.Environ
- pkg/credential/keygen: ed25519 SSH key generation helper used by onboard
- pkg/config: replace os.Getenv(PassphraseEnvVar) with
credential.PassphraseProvider() at all three call sites so that
LoadConfig and SaveConfig use whatever passphrase source is active
- cmd/picoclaw/onboard: prompt for passphrase with echo-off, generate
picoclaw-specific SSH key, re-encrypt existing config on re-onboard
- docs/credential_encryption.md: design doc for the enc:// format
* fix(credential): address Copilot review comments on PR #1521
- credential.go: decouple ErrPassphraseRequired from env var name;
message is now 'enc:// passphrase required' since PassphraseProvider
may come from any source, not just os.Environ
- credential.go: Resolver resolves symlinks via EvalSymlinks before the
isWithinDir containment check, preventing symlink-based path traversal
for file:// credential references
- store.go: tighten comment to describe only what SecureStore guarantees
(in-memory only); remove claims about how callers transport the value
- store_test.go: replace the meaningless GetReturnsCopy test (Go strings
are immutable, equality across two calls proves nothing) with
TestSecureStore_ConcurrentSetGet that exercises atomic.Pointer under
10-goroutine concurrent Set/Get load
- config_test.go: update error-message assertion to match new sentinel text
- docs/credential_encryption.md: remove reference to non-existent
'picoclaw encrypt' subcommand; describe the onboard flow instead
* fix(config): encryptPlaintextAPIKeys: struct-based encryption, fail-fast, remove raw []byte
* fix(credential): require SSH private key for encryption/decryption, remove passphrase-only mode
* lint: fix credential keygen lint, fix test keygen
* onboard: make encryption opt-in via --enc flag
Encryption (passphrase prompt + SSH key generation) is now only
triggered when the user passes --enc to 'picoclaw onboard'.
Without the flag, onboard skips the credential-encryption setup and
writes a plain config + workspace templates directly.
- Add --enc BoolFlag in NewOnboardCommand()
- Pass encrypt bool into onboard()
- Guard passphrase prompt, SSH key generation, and related env-var
setup behind the encrypt branch
- Adjust 'Next steps' output so the passphrase reminder only appears
when --enc was used
* fix: Use secure defaults for Pico channel setup and stop leaking the token in the URL
* fix: Derive default allow_origins from the setup request's Origin header instead of hardcoding localhost ports
* Add support for azure openai provider
* Add checks for deployment model name
* Apply suggestion from @Copilot
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* Addressing @Copilot suggestion to remove the init() function which seemed redundant
* Fix readme
* Fix linting checks
---------
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
* fix: safety guard incorrectly blocks commands with URLs
The absolutePathPattern regex was matching URL path components like
//github.com as file system paths, causing commands containing URLs
to be incorrectly blocked by the workspace restriction safety guard.
For example, 'agent-browser open https://github.com' would be blocked
because //github.com was treated as an absolute file path outside
the working directory.
The fix adds a check to skip any path match that starts with '//',
as these are URL path components, not file system paths.
Fixes#1203
* fix: handle file:// URIs correctly in safety guard
The previous fix skipped all paths starting with '//', which incorrectly
also skipped file:// URIs that could escape the workspace sandbox.
Changes:
- Only skip '//' paths when preceded by web URL schemes (http:, https:, ftp:, etc.)
- file:// URIs are now properly checked against workspace boundaries
- Added TestShellTool_FileURISandboxing to verify the fix
Fixes security issue raised by @alexhoshina in PR #1254
* style: fix gofumpt formatting
* fix(safety-guard): use exact match position to prevent URL exemption bypass
Using strings.Index(cmd, raw) always returned the first occurrence of the
matched substring, allowing a bypass where the same //path appeared both
inside a URL and as a standalone shell path (e.g. echo https://etc/passwd
&& cat //etc/passwd would skip the second match).
Switch to FindAllStringIndex so each match is evaluated at its actual
position in the command string.
Adds TestShellTool_URLBypassPrevented to cover the exploit scenario.
* make gateway aware of config.json change
* fix according to code review
* fix lint
* fix review comment
* fix for review
* refactor to fix review
* fix for review
* fix for review
* add model command to set default model
* fix for ci
* fix test for model
* fix active agent not recognized
* implement test for model command
* fix local-model can not set as default issue
* fix review comment
* fix for comment
* feat: add anthropic-messages protocol support
Add native Anthropic Messages API format support to enable
compatibility with custom endpoints that only support Anthropic's
native message format (not OpenAI-compatible format).
Changes:
- Add new pkg/providers/anthropic_messages package with HTTP-based provider
- Implement Anthropic Messages API request/response format conversion
- Add anthropic-messages protocol support in factory_provider.go
- Include comprehensive unit tests (64.2% coverage)
Features:
- Support for system, user, assistant, and tool messages
- Support for tool calls (tool_use blocks)
- Proper header handling (x-api-key, anthropic-version)
- Configurable max_tokens and temperature
- Automatic base URL normalization
Configuration example:
model: "anthropic-messages/claude-opus-4-6"
api_base: "https://api.anthropic.com"
api_key: "sk-..."
Tested with actual API endpoint, verified compatibility
with Anthropic Messages API specification.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* docs: add anthropic-messages protocol examples to README and config
Add configuration examples and documentation for the new
anthropic-messages protocol:
- config.example.json: Add claude-opus-4.6 example with anthropic-messages
- README.md: Add "Anthropic Messages API (native format)" section
- README.zh.md: Add Chinese version of the documentation
This helps users understand when to use anthropic-messages vs
anthropic protocol and fixes issue #269.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: format code with gofmt -s
- Align constant definitions in provider.go
- Align struct fields in test cases
- Fix gofmt formatting issues reported in review
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: address linter errors
- Fix HTTP header canonical form: "x-api-key" → "X-API-Key"
- Fix HTTP header canonical form: "anthropic-version" → "Anthropic-Version"
- Format imports with gci (standard, default, localmodule order)
- Format code with golines (max line length 120)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: resolve golangci-lint errors in anthropic-messages provider
- add nolint comment for canonicalheader rule on X-API-Key header (Anthropic API requires exact casing)
- fix golines formatting issues in provider_test.go (split long lines under 120 chars)
- fix long comment line in factory_provider.go (split into two lines)
Resolves CI linter failures for the anthropic-messages protocol implementation.
* fix(providers): address review comments in anthropic-messages provider
- fix normalizeBaseURL edge case that incorrectly appends /v1 to URLs already containing /v1 path (e.g., https://api.example.com/v1/proxy)
- remove dead code for apiBase empty check as normalizeBaseURL() always provides a default value
- update test to use proper constructor instead of direct struct initialization
- add detailed comments explaining the URL normalization logic
Resolves review comments on PR #1284
* fix(providers): remove hardcoded max_tokens in anthropic-messages provider
- remove hardcoded max_tokens value (4096) from buildRequestBody
- read max_tokens directly from options parameter
- add error handling when max_tokens is missing from options
- update test cases to include max_tokens in options
This fix ensures the provider respects the config default value (32768)
or system fallback (8192) instead of always using the hardcoded 4096.
* fix(providers): improve error handling and add edge case tests
- fix ToolCalls nil vs empty slice issue to ensure consistent JSON serialization
- add detailed HTTP error handling for common status codes (401, 429, 400, 404, 500, 503)
- add edge case tests for buildRequestBody and parseResponseBody
- clarify anthropic vs anthropic-messages protocol differences in docs
---------
Co-authored-by: Claude <noreply@anthropic.com>
When the claude CLI exits with a non-zero status, the previous error
handler only checked stderr. However, the CLI writes its output
(including error details) to stdout, especially when invoked with
--output-format json. This left the caller with only "exit status 1"
and no actionable information.
Now includes both stderr and stdout in the error message so the actual
failure reason is visible in logs.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(line): limit webhook request body size to prevent DoS
Add io.LimitReader with 1 MB cap on the LINE webhook handler to prevent
unauthenticated memory exhaustion via oversized POST requests.
Follows the same pattern used in the WeCom channel (io.LimitReader).
Requests exceeding the limit are rejected with 413 Request Entity Too Large.
Fixes#1407
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* refactor(line): hoist body size const, add boundary tests
- Move maxWebhookBodySize to package-level const
- Add TestWebhookAcceptsMaxBodySize (exact limit → 403, not 413)
- Add TestWebhookRejectsOversizedBodyBeforeSignatureCheck
- Use const in test instead of magic number
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* docs: swap header logo to webp, move meme logo to bottom
Replace header logo with assets/logo.webp across all 6 README
language variants and move the original meme logo (logo.jpg)
to the bottom of each file.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* docs: update GPT model names to gpt-5.4 and refine provider descriptions
Update all 6 language README variants:
- Correct GPT model references from gpt-5.2/gpt4 to gpt-5.4
- Refine provider descriptions in API Key comparison tables
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* chore: update default model to gpt-5.4, codex to gpt-5.3-codex
Update OpenAI default model references from gpt-5.2 to gpt-5.4
across source code, config examples, tests, and docs. Set Codex
default model to gpt-5.3-codex.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
TOOLS.md was intentionally removed in 21d60f6 and #771, as tools are
now provided to the LLM via JSON schema through ToProviderDefs().
These references were missed during that cleanup.
Suggested by @yinwm in #1355.
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* feat(providers): add LongCat model provider support
Add LongCat as an OpenAI-compatible provider with base URL
https://api.longcat.chat/openai and default model LongCat-Flash-Thinking.
Includes provider config, migration, factory routing, example config,
tests, and README entries for all 6 locales.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* fix(providers): address LongCat review feedback
- Add dedicated factory routing test for LongCat provider
- Add longcat to DefaultAPIBase test coverage
- Set default api_base in example config providers section
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
* test(providers): add ResolveProviderSelection tests for LongCat
Add two test cases to TestResolveProviderSelection:
- Explicit provider selection with api_base default and proxy wiring
- Fallback inference from model name with api_base default
---------
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Add UnmarshalText method to FlexibleStringSlice to support both English
(,) and Chinese (，) comma separators in environment variables.
Includes comprehensive unit tests covering:
- English commas, Chinese commas, mixed commas
- Single values, whitespace trimming
- Empty strings, edge cases
Fixes#1280
Non-OpenAI providers (Mistral, DeepSeek, Groq, etc.) reject unknown
request fields with 422 errors. The previous blocklist only excluded
Google/Gemini, but the comment already noted this feature is
OpenAI-only. Flip to an allowlist so only api.openai.com receives
the field.
Fixes#1333
- default tools.exec.allow_remote to true when omitted in config loading
- preserve allow_remote in OpenClaw config migration and API updates
- expose allow_remote in the web config form with i18n strings
- add backend and config tests covering the new default behavior
* feat(web_search): add load balance and failover for api keys
* feat(web_search): add load balance and failover for api keys
* lint
* new iter to get api key
* deleted conflicts
* feat(session): add SessionStore interface and JSONL backend adapter
Extract a SessionStore interface from the methods the agent loop uses
(AddMessage, GetHistory, SetSummary, TruncateHistory, Save, etc.).
Both SessionManager and the new JSONLBackend satisfy this interface,
allowing the persistence layer to be swapped transparently.
JSONLBackend wraps memory.Store and maps its error-returning API to
the fire-and-forget contract that the agent loop expects — write
errors are logged, reads return empty defaults on failure. Save()
triggers compaction to reclaim space after logical truncation.
Part of #1169
* test(session): add JSONLBackend integration tests
8 tests covering the full SessionStore contract through the JSONL
backend: message roundtrip, tool calls, summary, truncation with
compaction, history replacement, empty sessions, session isolation,
and the complete summarization flow (SetSummary → TruncateHistory →
Save).
Includes compile-time interface satisfaction checks for both
SessionManager and JSONLBackend.
Part of #1169
* feat(agent): wire JSONL session store into agent loop
Replace the concrete *SessionManager field with the SessionStore
interface and initialize the JSONL backend by default. Legacy .json
session files are auto-migrated on first startup. Falls back to
SessionManager if the JSONL store cannot be initialized.
The agent loop code (loop.go) requires zero changes — all method
calls work identically through the interface.
Closes#1169
* fix(session): propagate compact error from Save
Save() was swallowing the error returned by Compact and always
returning nil. Callers checking Save's return value would never
see a compaction failure. Return the error directly so the agent
loop can log or handle it as needed.
* feat(session): add Close to SessionStore interface
Add Close() error to SessionStore so callers can release resources
through the interface. JSONLBackend already had Close; this adds
a no-op implementation to SessionManager for compatibility.
* fix(session): close session stores on shutdown and harden migration
- Add Close() to AgentInstance, AgentRegistry, and AgentLoop so JSONL
file handles are released during gateway shutdown and CLI exit.
- Fall back to SessionManager when migration fails, preventing a split
state where some sessions live in JSONL and others remain in JSON.
- Add defer agentLoop.Close() in the CLI agent command path.
- Document SessionStore interface methods (fire-and-forget contract).
* feat(channels): enhance QQ channel with group support, typing, media, and URL sanitization
Add group message routing alongside existing C2C (direct) support using
chatType sync.Map to track whether a chatID is group or direct. Implement
passive reply with msg_id/msg_seq tracking for multi-part responses.
Add StartTyping (InputNotify msg_type=6 with periodic resend), SendMedia
(RichMediaMessage for HTTP/HTTPS URLs), and configurable Markdown message
support. Replace unbounded dedup map with TTL-based expiry and janitor
goroutine.
Sanitize URLs in group messages by replacing dots in domains with fullwidth
period to avoid QQ's URL blacklist rejection (error 40054010). Add rate
limit config (5 msg/s) and MaxMessageLength/SendMarkdown config fields.
* fix(channels): address review feedback on QQ channel implementation
- Fix goroutine leak: reinitialize done channel and sync.Once in Start()
to prevent multiple janitor goroutines on restart
- Fix double-close panic: guard close(done) with sync.Once in Stop()
- Fix StartTyping context: use c.ctx (channel lifecycle) instead of
caller's ctx (request lifecycle) for typing goroutine
- Refactor: extract getChatKind() helper to deduplicate chatType lookup
across Send(), StartTyping(), and SendMedia()
- Fix: use new(atomic.Uint64) instead of taking address of local var
- Fix: require explicit http(s):// scheme in URL regex to avoid false
positives on version strings like "1.2.3"
- Optimize: collect expired keys before deleting in dedupJanitor to
reduce lock hold time
- Fix: remove MaxMessageLength zero-value override in NewQQChannel
since defaults.go already sets 2000
* fix(channels): address second round of review feedback on QQ channel
- Fix SendMedia: bypass media store for direct http(s) URLs in part.Ref;
only fall back to store.Resolve for media:// refs; log clear warning
for local-only paths instead of silently skipping
- Fix chatType routing: default unknown chatIDs to "group" (safer for QQ
since outbound-only destinations like reasoning_channel_id are groups);
pre-register reasoning_channel_id as group at Start() time; add debug
log for untracked chatIDs
- Add dedup hard cap (10000 entries): evict oldest entry when map
exceeds capacity to prevent unbounded memory growth under high traffic
WithMaxMessageLength(4000) already ensures msg.Content ≤ 4000 chars
before reaching Send(), making the SplitMessage call redundant.
The HTML expansion safety net (re-split when >4096 after conversion)
is still preserved.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add separate User and RealName config fields (fall back to Nick)
- Make RequestCaps configurable (defaults to server-time, message-tags)
- Refactor isBotMentioned into nickMentionedAt returning position;
stripBotMention now uses nickMentionedAt internally
- Replace custom isAlphanumeric with unicode.IsLetter/unicode.IsDigit
- Update tests for new nickMentionedAt function
- Log job start with name, id, schedule kind, and channel
- Log job completion with duration and next run time
- Log job errors with duration and error message
- Helps diagnose scheduler stalls and connection issues
Fixes#1126
Go type assertions return true for zero values, which caused recurring
cron jobs (every_seconds/cron_expr) to silently become one-time 'at' tasks
when LLMs filled unused optional parameters with default values (0).
Changes:
- Add validity checks after type assertions: atSeconds > 0, everySeconds > 0, cronExpr != ""
- This ensures zero values are treated as 'not set' rather than valid schedule values
- Recurring tasks like "remind me every 2 hours" now correctly create recurring jobs
* feat(auth): add Anthropic OAuth setup-token login flow
Add support for Anthropic's OAuth-based setup tokens (sk-ant-oat01-*)
as an alternative to API keys. This includes:
- New `--setup-token` flag on `auth login` command
- Interactive login menu for Anthropic (setup token vs API key)
- Setup token validation and credential storage with oauth auth method
- Usage endpoint integration to show 5h/7d utilization in `auth status`
- Streaming support for OAuth tokens (required by Anthropic API)
- Model ID normalization (dots to hyphens) for API compatibility
- Remove .env.example (secrets should not be templated)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* feat(auth): update related functionality
* refactor(auth): organize constants and improve header casing in requests fo CI
* fix(auth): fix golint again
* fix(auth): handle nil arguments in tool calls for buildParams function
---------
Co-authored-by: Baller <sharonms3377@gmail.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* feat(feishu): implement SendMedia and add send_file tool
Add outbound media support for the Feishu channel so the agent can send
images and files to users via the MediaStore pipeline.
Feishu channel:
- SendMedia dispatches media parts as image or file uploads
- sendImage uploads via Image.Create then sends image message
- sendFile uploads via File.Create then sends file message
- feishuFileType maps extensions to Feishu file_type values
send_file tool:
- New tool lets the LLM send a local file to the current chat
- Validates path, registers file in MediaStore, returns media ref
- Agent loop wires tool registration, MediaStore propagation, and
context updates
Tested on Radxa Cubie A7A (arm64) with Feishu websocket channel.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(agent): publish outbound media regardless of SendResponse flag
The SendResponse flag controls whether the agent loop publishes the
final text response (callers that publish it themselves set this to
false). However, the media publish path was also gated behind this
flag, which meant tool-produced media was silently dropped for normal
channel messages.
Media should be published immediately when a tool returns media refs,
independent of how the text response is delivered.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(tools): use magic-bytes MIME detection and add file size limit to send_file
- Replace hardcoded extension-to-MIME map with h2non/filetype (magic
bytes) + mime.TypeByExtension fallback, consistent with the vision
pipeline in resolveMediaRefs
- Add configurable max file size check (defaults to config.DefaultMaxMediaSize,
20 MB) to prevent oversized uploads
- Add tests for magic-bytes detection, extension fallback, size limit,
and default max size
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* refactor(agent): add ForEachTool to AgentRegistry for cross-agent tool lookup
Extract the pattern of iterating agents to find a named tool into
AgentRegistry.ForEachTool, simplifying SetMediaStore propagation.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(agent,tools): adapt send_file to ctx-based channel injection after upstream refactor
Replace ContextualTool interface (removed upstream) with direct ctx
reading in SendFileTool.Execute, using ToolChannel/ToolChatID helpers.
Remove updateToolContexts which is no longer needed since ExecuteWithContext
already injects channel/chatID into ctx for all tools.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* feat(tools): support toggling send_file tool via config
Add SendFileConfig with Enabled field to ToolsConfig, defaulting to
true. Wrap send_file tool registration in loop.go with the config
check, consistent with the pattern used by other tools.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* feat(commands): Session management [Phase 1/2] command centralization and registration
* docs: add design for command registry post-review fixes
Documents the architecture decisions for fixing 5 Important issues
from code review: SubCommand pattern, Deps struct, command-group files,
Executor caching, and Telegram registration dedup.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* feat(commands): add SubCommand type and EffectiveUsage method
Introduce SubCommand struct for declaring sub-commands structurally
within a parent command Definition. The EffectiveUsage() method
auto-generates usage strings from sub-command names and args,
preventing drift between help text and actual handler behavior.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* feat(commands): add Deps struct and secondToken helper, remove dead contains()
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* feat(commands): add sub-command routing to Executor
Uses Registry.Lookup for O(1) command dispatch instead of iterating
all definitions. Definitions with SubCommands are routed to matching
sub-command handlers. Missing or unknown sub-commands reply with
auto-generated usage.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* refactor(commands): split into command-group files with Deps injection
Extract show/list/start/help into individual cmd_*.go files.
Replace config.Config parameter with Deps struct for runtime data.
Restore /show agents and /list agents sub-commands.
Use EffectiveUsage for auto-generated help text.
Bridge external callers (agent/loop.go, telegram.go) with Deps wrapper
until Task 5 fully wires the Deps fields.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* perf(commands): cache Executor in AgentLoop, wire Deps with runtime callbacks
Create Executor once in NewAgentLoop instead of per-message. Deps
closures capture AgentLoop pointer for late-bound access to
channelManager and runtime agent model.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(telegram): remove duplicate initBotCommands, keep async startCommandRegistration only
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* chore(commands): restore Outcome comments and annotate Deps.Config
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* refactor(commands): consolidate /switch into commands package, fix ! prefix
Move /switch model and /switch channel handling from inline loop.go
logic into cmd_switch.go using the SubCommand + Deps pattern. This
removes the OutcomePassthrough branch in handleCommand entirely.
Also replace the hardcoded "/" prefix check with commands.HasCommandPrefix
so that "!" prefixed commands are correctly routed to the Executor.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* chore: add docs/plans to .gitignore and untrack existing files
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* refactor(commands): address code review findings
- Remove dead ExecuteResult.Reply field and unused branch in loop.go
- Extract shared agentsHandler for /show agents and /list agents
- Remove redundant firstToken/secondToken (use nthToken instead)
- Simplify Telegram startup: pass BuiltinDefinitions directly
- Centralize req.Reply nil guard in executeDefinition
- Extract unavailableMsg constant (was duplicated 5 times)
- Remove unused MessageID from Request
- Remove stale "reserved for Phase 2" comment on Deps.Config
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* refactor(commands): replace Deps with per-request Runtime
Separate stateless Registry (cached on AgentLoop) from per-request
Runtime (passed to handlers at execution time). This enables future
session management features to inject per-request context without
modifying the command registry.
- Rename Deps → Runtime, move to runtime.go
- Change Handler signature: func(ctx, req) error → func(ctx, req, rt *Runtime) error
- NewExecutor now takes (registry, runtime) — executor is created per-request
- BuiltinDefinitions() no longer takes parameters (stateless)
- AgentLoop caches cmdRegistry, builds Runtime via buildRuntime()
- Update all cmd_*.go handlers and tests
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* style: fix gci import grouping and godoc formatting
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(onboard): skip legacy AGENT.md when copying embedded workspace templates
The workspace/ directory contains both AGENT.md (legacy) and AGENTS.md
(current). copyEmbeddedToTarget was copying both, causing the test
TestCopyEmbeddedToTargetUsesAgentsMarkdown to fail. Skip AGENT.md
during the walk to match the expected behavior.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* refactor(agent): address self-review comments on loop.go
- Move cmdRegistry init into struct literal (review comment #11)
- Rename buildRuntime → buildCommandsRuntime for clarity (review comment #12)
- Add comment to default switch case explaining passthrough (review comment #13)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* refactor(commands): address code review findings on naming and correctness
- Rename dispatcher.go → request.go (no Dispatcher type remains)
- Rename cmd_agents.go → handler_agents.go (shared handler, not a top-level command)
- Add modelMu to protect AgentInstance.Model writes in SwitchModel
- Add ListDefinitions to Runtime so /help uses registry instead of BuiltinDefinitions()
- Fix SwitchChannel message: validation-only callback should not say "Switched"
- Propagate Reply errors in executor instead of discarding with _ =
- Add HasCommandPrefix unit test
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* refactor(onboard): extract legacy filename to constant
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(agent): handle commands before route error check
Move handleCommand() before the routeErr gate so global commands
(/help, /show, /switch) remain available even when routing fails.
Context-dependent commands that need a routed agent will report
"unavailable" through their nil-Runtime guards.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* revert: remove unnecessary AGENT.md skip in onboard
Reverts 02d0c04 and 74deae1. The test failure was caused by a local
leftover workspace/AGENT.md file (gitignored but embedded by go:embed).
Deleting the local file fixes the root cause; the code-level skip was
never needed.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix: executeDefinition Unknown option
* fix(agent): use routed agent for model commands, restore Telegram command diff
- Remove modelMu: message processing is serial, no concurrent writes
- Pass routed agent to handleCommand/buildCommandsRuntime instead of
always using default agent
- GetModelInfo/SwitchModel are nil when agent is nil (route failed),
handlers reply "unavailable"
- Restore GetMyCommands + slices.Equal check before SetMyCommands to
avoid unnecessary Telegram API calls on restart
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* fix(commands): remove unintended config mutation in SwitchModel
SwitchModel should only update the routed agent's runtime Model field.
Writing to cfg.Agents.Defaults.ModelName was a behavioral change that
corrupts the default agent config when switching a non-default agent.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* refactor(commands): move /switch channel to /check channel
/switch channel only validates availability, not actually switching.
Rename to /check channel to match actual behavior. /switch channel
now shows a redirect message pointing users to the new command.
Addresses review feedback from yinwm on PR #959.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
1. CJK token estimation: replace flat rune_count/3 with script-aware
counting — CJK runes (U+2E80–U+9FFF, U+F900–U+FAFF, U+AC00–U+D7AF)
count as 1 token each, non-CJK runes at /4. This fixes a 3x
underestimate for Chinese/Japanese/Korean text that could incorrectly
route complex CJK messages to the light model.
2. Routing observability: SelectModel now returns the computed score as
a third value. selectCandidates logs the score on both paths — Info
level for light model selection, Debug level for primary model
selection.
3. Added tests: TestExtractFeatures_TokenEstimate_Mixed (CJK+ASCII mix),
TestRouter_SelectModel_ReturnsScore.
Addresses review feedback from @mingmxren.
Upstream added ThinkingLevel, SummarizeMessageThreshold,
SummarizeTokenPercent, MaxMediaSize, and maybeSummarize.
Our branch added Router, LightCandidates, and selectCandidates.
Both sets of changes are kept. Dead updateToolContexts removed
(upstream deleted it; no callers exist).
Add TimeoutSeconds field to ExecConfig so the shell command execution
timeout can be configured instead of being hardcoded to 60s.
- Add TimeoutSeconds int field to ExecConfig in pkg/config/config.go
with json/env tags (PICOCLAW_TOOLS_EXEC_TIMEOUT_SECONDS)
- Set default value of 60s in DefaultConfig() in pkg/config/defaults.go
- Read TimeoutSeconds from config in NewExecToolWithConfig() in
pkg/tools/shell.go; falls back to 60s when value is 0 or unset
Sort irc import alphabetically in helpers.go and fix struct field
alignment in irc.go to satisfy golangci-lint gci formatter.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
io.ReadAll errors were silently discarded with `body, _ := io.ReadAll(...)`,
which could cause empty or partial data to be used for JSON unmarshaling
or error messages. This adds proper error checks for all instances.
Add IRC as a new channel for picoclaw, supporting server connections,
channel joins, DMs, mention-based group triggers, and IRCv3 typing
indicators. Uses ergochat/irc-go for connection management with SASL,
NickServ, and automatic reconnection support.
Closes#1137
* fix: eliminate data races on shared tool instances
Signed-off-by: Boris Bliznioukov <blib@mail.com>
* fix: remove unused indirect dependency on github.com/gdamore/tcell/v2
Signed-off-by: Boris Bliznioukov <blib@mail.com>
* fix: reviewer comments improve context handling for tool execution and ensure defaults for non-conversation callers
Signed-off-by: Boris Bliznioukov <blib@mail.com>
---------
Signed-off-by: Boris Bliznioukov <blib@mail.com>
* feat: add extended thinking support for Anthropic models
Support configurable thinking levels (off/low/medium/high/xhigh/adaptive)
via `agents.defaults.thinking_level` config field.
- "adaptive": uses Anthropic's adaptive thinking API (Claude 4.6+)
- "low/medium/high/xhigh": uses budget_tokens (all thinking-capable models)
- "off": disables thinking (default)
API constraints handled:
- Temperature cleared when thinking is enabled
- budget_tokens clamped to max_tokens-1
- Thinking response blocks parsed into Reasoning field
Relates to #645, #966
* fix: address PR review feedback for thinking support
- Add ThinkingCapable interface for provider capability detection
- Warn when thinking_level is set but provider doesn't support it
- Warn when temperature is cleared due to thinking enabled
- Adjust budget values per Anthropic best practices (medium=16K, xhigh=64K)
- Add budget clamp warning and 80% threshold warning
- Add parseResponse thinking block tests
- Add thinking_level field to config.example.json
* refactor: move ThinkingLevel from AgentDefaults to ModelConfig
Thinking is a model-level capability, not a global agent property.
Per-model config avoids silent ignoring on non-Anthropic providers
and eliminates spurious warning logs in multi-provider setups.
Addresses PR #1076 review feedback from @yinwm.
Resolve merge conflicts to keep both SearXNG and GLM Search
providers. Updated search priority order to:
Perplexity > Brave > SearXNG > Tavily > DuckDuckGo > GLM Search
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The TestConvertProvidersToModelList_AllProviders test expected 19
providers but adding Avian brings the total to 20.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Accept upstream versions for all non-Telegram files to keep PR
scope focused on Telegram message chunking only.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
