2026-03-13 06:20:24 +00:00
|
|
|
// PicoClaw - Ultra-lightweight personal AI agent
|
|
|
|
|
// License: MIT
|
|
|
|
|
//
|
|
|
|
|
// Copyright (c) 2026 PicoClaw contributors
|
|
|
|
|
|
|
|
|
|
package agent
|
|
|
|
|
|
|
|
|
|
import (
|
|
|
|
|
"github.com/sipeed/picoclaw/pkg/providers"
|
feat(seahorse): implement short-term memory engine (LCM) (#2285)
* feat(seahorse): implement short-term memory engine of seahorse
Add pkg/seahorse/ module implementing a SQLite-backed DAG-based summary
hierarchy for context management, ported from lossless-claw's LCM design:
- types.go + short_constants.go: core types (Message, Summary, Conversation,
ContextItem) and configuration constants (fanout, token targets, thresholds)
- migration.go: idempotent DB schema with FTS5 trigram tokenizer for CJK
- store.go: full SQLite CRUD (conversations, messages, summaries DAG,
context_items with ordinal gap numbering, FTS5 search)
- short_engine.go: Engine lifecycle (NewEngine, Ingest, Assemble, Compact),
session pattern filtering (ignore/stateless glob→regex compilation),
per-session mutex via sync.Map
- short_assembler.go: budget-aware context assembly with fresh tail protection
(32 messages), oldest-first eviction, summary XML formatting, RebuildContextItems
- short_compaction.go: leaf compaction (messages→summary) and condensed
compaction (summaries→higher-level summary), 3-level LLM escalation,
CompactUntilUnder for emergency overflow
- short_retrieval.go: lookupByID, FTS5/LIKE search, recursive expand with
token cap
- context_seahorse.go: agent.ContextManager adapter, registered as "seahorse",
provider↔seahorse message type conversion (ToolCalls, tool_result)
* fix(seahorse): correct 3 adapter bugs in context management
- TokenCount: use full message (Content+ToolCalls+Media) instead of Content-only
- Empty Content: rebuild Content from tool_result Parts when stored empty
- Duplicate summaries: summaries only in Summary field, not in History messages
- Grep: fix SearchResult.Snippet→Content for summaries
- Schema: fix FTS5 SQL uses VIRTUAL TABLE not TEMP TABLE
- TestFTS5SQLConstants: verify FTS5 SQL syntax correctness
- Test: fix flaky TestCompactLeaf
* fix(agent): ingest steering messages into seahorse SQLite
Steering messages were only persisted to session JSONL but not ingested
into seahorse SQLite, causing them to be missing from context assembly.
Added `ts.ingestMessage(turnCtx, al, pm)` call in the steering message
injection block alongside the existing JSONL persistence.
Test: TestSeahorseSteeringMessageIngested verifies steering messages
appear in seahorse SQLite DB after being processed.
* fix(seahorse): address 3 blocking bugs from code review
- Fix resequenceContextItemsTx scan error handling (store.go:850)
Changed `return err` to `return scanErr` to properly propagate scan errors
instead of returning nil (which silently corrupts data)
- Fix sql.NullString for INTEGER column (store.go:847)
Changed `mid` from sql.NullString to sql.NullInt64 since message_id
is INTEGER in schema. Removed unnecessary strconv.ParseInt call.
- Fix compactCondensed fallback deleting non-candidate items
Added ReplaceContextItemsWithSummary method for per-item deletion
when candidates are not contiguous in ordinal space.
Optimized to use range deletion when candidates are consecutive.
* fix(seahorse): pass Budget to Compact for correct condensed threshold
Issue #4 from PR review: When Budget was not passed to seahorse.Compact,
it defaulted to `tokensBefore * 0.75`, making `tokensBefore > budget`
always true and causing condensed compaction to trigger unnecessarily.
Changes:
- context_seahorse.go: Forward Budget from CompactRequest to CompactInput
- loop.go: Pass Budget (ContextWindow) in all 3 Compact calls
- Add test verifying condensed is skipped when tokens < threshold
- Fix lint issues in store.go and store_test.go
* fix(seahorse): add mutex for assembler lazy initialization
Issue #5 from PR review: The check-then-create pattern for e.assembler
was a data race when multiple goroutines called Assemble() concurrently:
if e.assembler == nil {
e.assembler = &Assembler{...}
}
Changes:
- Add assemblerMu sync.Mutex to Engine struct
- Add initAssemblerOnce() using double-checked locking (same pattern as initCompactionOnce)
- Add TestAssemblerLazyInitRace to verify thread-safety
* fix(seahorse): handle non-consecutive depths in selectShallowestCondensationCandidate
Issue #8 from PR review: the loop iterated depth 0, 1, 2... assuming
consecutive keys, but break when key was missing caused deeper depths
to never be checked.
Fix: collect all existing depth keys, sort, then iterate in order.
* fix(seahorse): wrap DeleteMessagesAfterID and appendContextItems in transactions
- DeleteMessagesAfterID: wrap all DELETE operations in a transaction for
atomicity, remove redundant manual FTS delete (handled by trigger)
- appendContextItems: use transaction to fix read-then-write race condition
- Add GetMaxOrdinalTx and resolveItemTokenCountTx for transaction-scoped queries
- Remove unused resolveItemTokenCount function
Fixes PR review issues 6 and 7.
* fix(seahorse): derive readable content from Parts and cap CompactUntilUnder iterations
- Derive readable content from MessageParts in AddMessageWithParts so
FTS5 indexing and summary formatting can access tool call information
- formatMessagesForSummary and truncateSummary now fall back to Parts
when Content is empty, fixing blank summaries for Part-based messages
- Add MaxCompactIterations (20) to prevent CompactUntilUnder infinite
loops; exceeded iterations are logged as warnings
2026-04-05 01:05:16 +00:00
|
|
|
"github.com/sipeed/picoclaw/pkg/tokenizer"
|
2026-03-13 06:20:24 +00:00
|
|
|
)
|
|
|
|
|
|
2026-03-13 07:54:50 +00:00
|
|
|
// parseTurnBoundaries returns the starting index of each Turn in the history.
|
|
|
|
|
// A Turn is a complete "user input → LLM iterations → final response" cycle
|
|
|
|
|
// (as defined in #1316). Each Turn begins at a user message and extends
|
|
|
|
|
// through all subsequent assistant/tool messages until the next user message.
|
2026-03-13 06:20:24 +00:00
|
|
|
//
|
2026-03-13 07:54:50 +00:00
|
|
|
// Cutting at a Turn boundary guarantees that no tool-call sequence
|
|
|
|
|
// (assistant+ToolCalls → tool results) is split across the cut.
|
|
|
|
|
func parseTurnBoundaries(history []providers.Message) []int {
|
|
|
|
|
var starts []int
|
|
|
|
|
for i, msg := range history {
|
|
|
|
|
if msg.Role == "user" {
|
|
|
|
|
starts = append(starts, i)
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
return starts
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
// isSafeBoundary reports whether index is a valid Turn boundary — i.e.,
|
|
|
|
|
// a position where the kept portion (history[index:]) begins at a user
|
|
|
|
|
// message, so no tool-call sequence is torn apart.
|
2026-03-13 06:20:24 +00:00
|
|
|
func isSafeBoundary(history []providers.Message, index int) bool {
|
|
|
|
|
if index <= 0 || index >= len(history) {
|
|
|
|
|
return true
|
|
|
|
|
}
|
|
|
|
|
return history[index].Role == "user"
|
|
|
|
|
}
|
|
|
|
|
|
2026-03-13 07:54:50 +00:00
|
|
|
// findSafeBoundary locates the nearest Turn boundary to targetIndex.
|
|
|
|
|
// It prefers the boundary at or before targetIndex (preserving more recent
|
|
|
|
|
// context). Falls back to the nearest boundary after targetIndex, and
|
|
|
|
|
// returns targetIndex unchanged only when no Turn boundary exists at all.
|
2026-03-13 06:20:24 +00:00
|
|
|
func findSafeBoundary(history []providers.Message, targetIndex int) int {
|
|
|
|
|
if len(history) == 0 {
|
|
|
|
|
return 0
|
|
|
|
|
}
|
|
|
|
|
if targetIndex <= 0 {
|
|
|
|
|
return 0
|
|
|
|
|
}
|
|
|
|
|
if targetIndex >= len(history) {
|
|
|
|
|
return len(history)
|
|
|
|
|
}
|
|
|
|
|
|
2026-03-13 07:54:50 +00:00
|
|
|
turns := parseTurnBoundaries(history)
|
|
|
|
|
if len(turns) == 0 {
|
2026-03-13 06:20:24 +00:00
|
|
|
return targetIndex
|
|
|
|
|
}
|
|
|
|
|
|
2026-03-13 07:54:50 +00:00
|
|
|
// Find the last Turn boundary at or before targetIndex.
|
|
|
|
|
// Prefer backward: keeps more recent messages.
|
|
|
|
|
backward := -1
|
|
|
|
|
for _, t := range turns {
|
|
|
|
|
if t <= targetIndex {
|
|
|
|
|
backward = t
|
2026-03-13 06:20:24 +00:00
|
|
|
}
|
|
|
|
|
}
|
2026-03-13 07:54:50 +00:00
|
|
|
if backward > 0 {
|
|
|
|
|
return backward
|
|
|
|
|
}
|
2026-03-13 06:20:24 +00:00
|
|
|
|
2026-03-13 07:54:50 +00:00
|
|
|
// No valid Turn boundary before target (or only at index 0 which
|
|
|
|
|
// would keep everything). Use the first Turn after targetIndex.
|
|
|
|
|
for _, t := range turns {
|
|
|
|
|
if t > targetIndex {
|
|
|
|
|
return t
|
2026-03-13 06:20:24 +00:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2026-03-13 08:25:27 +00:00
|
|
|
// No Turn boundary after targetIndex either. The only boundary is at
|
|
|
|
|
// index 0, meaning the entire history is a single Turn. Return 0 to
|
|
|
|
|
// signal that safe compression is not possible — callers check for
|
|
|
|
|
// mid <= 0 and skip compression in that case.
|
|
|
|
|
return 0
|
2026-03-13 06:20:24 +00:00
|
|
|
}
|
|
|
|
|
|
feat(seahorse): implement short-term memory engine (LCM) (#2285)
* feat(seahorse): implement short-term memory engine of seahorse
Add pkg/seahorse/ module implementing a SQLite-backed DAG-based summary
hierarchy for context management, ported from lossless-claw's LCM design:
- types.go + short_constants.go: core types (Message, Summary, Conversation,
ContextItem) and configuration constants (fanout, token targets, thresholds)
- migration.go: idempotent DB schema with FTS5 trigram tokenizer for CJK
- store.go: full SQLite CRUD (conversations, messages, summaries DAG,
context_items with ordinal gap numbering, FTS5 search)
- short_engine.go: Engine lifecycle (NewEngine, Ingest, Assemble, Compact),
session pattern filtering (ignore/stateless glob→regex compilation),
per-session mutex via sync.Map
- short_assembler.go: budget-aware context assembly with fresh tail protection
(32 messages), oldest-first eviction, summary XML formatting, RebuildContextItems
- short_compaction.go: leaf compaction (messages→summary) and condensed
compaction (summaries→higher-level summary), 3-level LLM escalation,
CompactUntilUnder for emergency overflow
- short_retrieval.go: lookupByID, FTS5/LIKE search, recursive expand with
token cap
- context_seahorse.go: agent.ContextManager adapter, registered as "seahorse",
provider↔seahorse message type conversion (ToolCalls, tool_result)
* fix(seahorse): correct 3 adapter bugs in context management
- TokenCount: use full message (Content+ToolCalls+Media) instead of Content-only
- Empty Content: rebuild Content from tool_result Parts when stored empty
- Duplicate summaries: summaries only in Summary field, not in History messages
- Grep: fix SearchResult.Snippet→Content for summaries
- Schema: fix FTS5 SQL uses VIRTUAL TABLE not TEMP TABLE
- TestFTS5SQLConstants: verify FTS5 SQL syntax correctness
- Test: fix flaky TestCompactLeaf
* fix(agent): ingest steering messages into seahorse SQLite
Steering messages were only persisted to session JSONL but not ingested
into seahorse SQLite, causing them to be missing from context assembly.
Added `ts.ingestMessage(turnCtx, al, pm)` call in the steering message
injection block alongside the existing JSONL persistence.
Test: TestSeahorseSteeringMessageIngested verifies steering messages
appear in seahorse SQLite DB after being processed.
* fix(seahorse): address 3 blocking bugs from code review
- Fix resequenceContextItemsTx scan error handling (store.go:850)
Changed `return err` to `return scanErr` to properly propagate scan errors
instead of returning nil (which silently corrupts data)
- Fix sql.NullString for INTEGER column (store.go:847)
Changed `mid` from sql.NullString to sql.NullInt64 since message_id
is INTEGER in schema. Removed unnecessary strconv.ParseInt call.
- Fix compactCondensed fallback deleting non-candidate items
Added ReplaceContextItemsWithSummary method for per-item deletion
when candidates are not contiguous in ordinal space.
Optimized to use range deletion when candidates are consecutive.
* fix(seahorse): pass Budget to Compact for correct condensed threshold
Issue #4 from PR review: When Budget was not passed to seahorse.Compact,
it defaulted to `tokensBefore * 0.75`, making `tokensBefore > budget`
always true and causing condensed compaction to trigger unnecessarily.
Changes:
- context_seahorse.go: Forward Budget from CompactRequest to CompactInput
- loop.go: Pass Budget (ContextWindow) in all 3 Compact calls
- Add test verifying condensed is skipped when tokens < threshold
- Fix lint issues in store.go and store_test.go
* fix(seahorse): add mutex for assembler lazy initialization
Issue #5 from PR review: The check-then-create pattern for e.assembler
was a data race when multiple goroutines called Assemble() concurrently:
if e.assembler == nil {
e.assembler = &Assembler{...}
}
Changes:
- Add assemblerMu sync.Mutex to Engine struct
- Add initAssemblerOnce() using double-checked locking (same pattern as initCompactionOnce)
- Add TestAssemblerLazyInitRace to verify thread-safety
* fix(seahorse): handle non-consecutive depths in selectShallowestCondensationCandidate
Issue #8 from PR review: the loop iterated depth 0, 1, 2... assuming
consecutive keys, but break when key was missing caused deeper depths
to never be checked.
Fix: collect all existing depth keys, sort, then iterate in order.
* fix(seahorse): wrap DeleteMessagesAfterID and appendContextItems in transactions
- DeleteMessagesAfterID: wrap all DELETE operations in a transaction for
atomicity, remove redundant manual FTS delete (handled by trigger)
- appendContextItems: use transaction to fix read-then-write race condition
- Add GetMaxOrdinalTx and resolveItemTokenCountTx for transaction-scoped queries
- Remove unused resolveItemTokenCount function
Fixes PR review issues 6 and 7.
* fix(seahorse): derive readable content from Parts and cap CompactUntilUnder iterations
- Derive readable content from MessageParts in AddMessageWithParts so
FTS5 indexing and summary formatting can access tool call information
- formatMessagesForSummary and truncateSummary now fall back to Parts
when Content is empty, fixing blank summaries for Part-based messages
- Add MaxCompactIterations (20) to prevent CompactUntilUnder infinite
loops; exceeded iterations are logged as warnings
2026-04-05 01:05:16 +00:00
|
|
|
// EstimateMessageTokens estimates the token count for a single message.
|
|
|
|
|
// Delegates to the shared tokenizer package for consistency across agent and seahorse.
|
|
|
|
|
func EstimateMessageTokens(msg providers.Message) int {
|
|
|
|
|
return tokenizer.EstimateMessageTokens(msg)
|
2026-03-13 06:20:24 +00:00
|
|
|
}
|
|
|
|
|
|
feat(seahorse): implement short-term memory engine (LCM) (#2285)
* feat(seahorse): implement short-term memory engine of seahorse
Add pkg/seahorse/ module implementing a SQLite-backed DAG-based summary
hierarchy for context management, ported from lossless-claw's LCM design:
- types.go + short_constants.go: core types (Message, Summary, Conversation,
ContextItem) and configuration constants (fanout, token targets, thresholds)
- migration.go: idempotent DB schema with FTS5 trigram tokenizer for CJK
- store.go: full SQLite CRUD (conversations, messages, summaries DAG,
context_items with ordinal gap numbering, FTS5 search)
- short_engine.go: Engine lifecycle (NewEngine, Ingest, Assemble, Compact),
session pattern filtering (ignore/stateless glob→regex compilation),
per-session mutex via sync.Map
- short_assembler.go: budget-aware context assembly with fresh tail protection
(32 messages), oldest-first eviction, summary XML formatting, RebuildContextItems
- short_compaction.go: leaf compaction (messages→summary) and condensed
compaction (summaries→higher-level summary), 3-level LLM escalation,
CompactUntilUnder for emergency overflow
- short_retrieval.go: lookupByID, FTS5/LIKE search, recursive expand with
token cap
- context_seahorse.go: agent.ContextManager adapter, registered as "seahorse",
provider↔seahorse message type conversion (ToolCalls, tool_result)
* fix(seahorse): correct 3 adapter bugs in context management
- TokenCount: use full message (Content+ToolCalls+Media) instead of Content-only
- Empty Content: rebuild Content from tool_result Parts when stored empty
- Duplicate summaries: summaries only in Summary field, not in History messages
- Grep: fix SearchResult.Snippet→Content for summaries
- Schema: fix FTS5 SQL uses VIRTUAL TABLE not TEMP TABLE
- TestFTS5SQLConstants: verify FTS5 SQL syntax correctness
- Test: fix flaky TestCompactLeaf
* fix(agent): ingest steering messages into seahorse SQLite
Steering messages were only persisted to session JSONL but not ingested
into seahorse SQLite, causing them to be missing from context assembly.
Added `ts.ingestMessage(turnCtx, al, pm)` call in the steering message
injection block alongside the existing JSONL persistence.
Test: TestSeahorseSteeringMessageIngested verifies steering messages
appear in seahorse SQLite DB after being processed.
* fix(seahorse): address 3 blocking bugs from code review
- Fix resequenceContextItemsTx scan error handling (store.go:850)
Changed `return err` to `return scanErr` to properly propagate scan errors
instead of returning nil (which silently corrupts data)
- Fix sql.NullString for INTEGER column (store.go:847)
Changed `mid` from sql.NullString to sql.NullInt64 since message_id
is INTEGER in schema. Removed unnecessary strconv.ParseInt call.
- Fix compactCondensed fallback deleting non-candidate items
Added ReplaceContextItemsWithSummary method for per-item deletion
when candidates are not contiguous in ordinal space.
Optimized to use range deletion when candidates are consecutive.
* fix(seahorse): pass Budget to Compact for correct condensed threshold
Issue #4 from PR review: When Budget was not passed to seahorse.Compact,
it defaulted to `tokensBefore * 0.75`, making `tokensBefore > budget`
always true and causing condensed compaction to trigger unnecessarily.
Changes:
- context_seahorse.go: Forward Budget from CompactRequest to CompactInput
- loop.go: Pass Budget (ContextWindow) in all 3 Compact calls
- Add test verifying condensed is skipped when tokens < threshold
- Fix lint issues in store.go and store_test.go
* fix(seahorse): add mutex for assembler lazy initialization
Issue #5 from PR review: The check-then-create pattern for e.assembler
was a data race when multiple goroutines called Assemble() concurrently:
if e.assembler == nil {
e.assembler = &Assembler{...}
}
Changes:
- Add assemblerMu sync.Mutex to Engine struct
- Add initAssemblerOnce() using double-checked locking (same pattern as initCompactionOnce)
- Add TestAssemblerLazyInitRace to verify thread-safety
* fix(seahorse): handle non-consecutive depths in selectShallowestCondensationCandidate
Issue #8 from PR review: the loop iterated depth 0, 1, 2... assuming
consecutive keys, but break when key was missing caused deeper depths
to never be checked.
Fix: collect all existing depth keys, sort, then iterate in order.
* fix(seahorse): wrap DeleteMessagesAfterID and appendContextItems in transactions
- DeleteMessagesAfterID: wrap all DELETE operations in a transaction for
atomicity, remove redundant manual FTS delete (handled by trigger)
- appendContextItems: use transaction to fix read-then-write race condition
- Add GetMaxOrdinalTx and resolveItemTokenCountTx for transaction-scoped queries
- Remove unused resolveItemTokenCount function
Fixes PR review issues 6 and 7.
* fix(seahorse): derive readable content from Parts and cap CompactUntilUnder iterations
- Derive readable content from MessageParts in AddMessageWithParts so
FTS5 indexing and summary formatting can access tool call information
- formatMessagesForSummary and truncateSummary now fall back to Parts
when Content is empty, fixing blank summaries for Part-based messages
- Add MaxCompactIterations (20) to prevent CompactUntilUnder infinite
loops; exceeded iterations are logged as warnings
2026-04-05 01:05:16 +00:00
|
|
|
// EstimateToolDefsTokens estimates the total token cost of tool definitions
|
|
|
|
|
// as they appear in the LLM request. Delegates to the shared tokenizer package.
|
|
|
|
|
func EstimateToolDefsTokens(defs []providers.ToolDefinition) int {
|
|
|
|
|
return tokenizer.EstimateToolDefsTokens(defs)
|
2026-03-13 06:20:24 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
// isOverContextBudget checks whether the assembled messages plus tool definitions
|
|
|
|
|
// and output reserve would exceed the model's context window. This enables
|
|
|
|
|
// proactive compression before calling the LLM, rather than reacting to 400 errors.
|
|
|
|
|
func isOverContextBudget(
|
|
|
|
|
contextWindow int,
|
|
|
|
|
messages []providers.Message,
|
|
|
|
|
toolDefs []providers.ToolDefinition,
|
|
|
|
|
maxTokens int,
|
|
|
|
|
) bool {
|
|
|
|
|
msgTokens := 0
|
|
|
|
|
for _, m := range messages {
|
feat(seahorse): implement short-term memory engine (LCM) (#2285)
* feat(seahorse): implement short-term memory engine of seahorse
Add pkg/seahorse/ module implementing a SQLite-backed DAG-based summary
hierarchy for context management, ported from lossless-claw's LCM design:
- types.go + short_constants.go: core types (Message, Summary, Conversation,
ContextItem) and configuration constants (fanout, token targets, thresholds)
- migration.go: idempotent DB schema with FTS5 trigram tokenizer for CJK
- store.go: full SQLite CRUD (conversations, messages, summaries DAG,
context_items with ordinal gap numbering, FTS5 search)
- short_engine.go: Engine lifecycle (NewEngine, Ingest, Assemble, Compact),
session pattern filtering (ignore/stateless glob→regex compilation),
per-session mutex via sync.Map
- short_assembler.go: budget-aware context assembly with fresh tail protection
(32 messages), oldest-first eviction, summary XML formatting, RebuildContextItems
- short_compaction.go: leaf compaction (messages→summary) and condensed
compaction (summaries→higher-level summary), 3-level LLM escalation,
CompactUntilUnder for emergency overflow
- short_retrieval.go: lookupByID, FTS5/LIKE search, recursive expand with
token cap
- context_seahorse.go: agent.ContextManager adapter, registered as "seahorse",
provider↔seahorse message type conversion (ToolCalls, tool_result)
* fix(seahorse): correct 3 adapter bugs in context management
- TokenCount: use full message (Content+ToolCalls+Media) instead of Content-only
- Empty Content: rebuild Content from tool_result Parts when stored empty
- Duplicate summaries: summaries only in Summary field, not in History messages
- Grep: fix SearchResult.Snippet→Content for summaries
- Schema: fix FTS5 SQL uses VIRTUAL TABLE not TEMP TABLE
- TestFTS5SQLConstants: verify FTS5 SQL syntax correctness
- Test: fix flaky TestCompactLeaf
* fix(agent): ingest steering messages into seahorse SQLite
Steering messages were only persisted to session JSONL but not ingested
into seahorse SQLite, causing them to be missing from context assembly.
Added `ts.ingestMessage(turnCtx, al, pm)` call in the steering message
injection block alongside the existing JSONL persistence.
Test: TestSeahorseSteeringMessageIngested verifies steering messages
appear in seahorse SQLite DB after being processed.
* fix(seahorse): address 3 blocking bugs from code review
- Fix resequenceContextItemsTx scan error handling (store.go:850)
Changed `return err` to `return scanErr` to properly propagate scan errors
instead of returning nil (which silently corrupts data)
- Fix sql.NullString for INTEGER column (store.go:847)
Changed `mid` from sql.NullString to sql.NullInt64 since message_id
is INTEGER in schema. Removed unnecessary strconv.ParseInt call.
- Fix compactCondensed fallback deleting non-candidate items
Added ReplaceContextItemsWithSummary method for per-item deletion
when candidates are not contiguous in ordinal space.
Optimized to use range deletion when candidates are consecutive.
* fix(seahorse): pass Budget to Compact for correct condensed threshold
Issue #4 from PR review: When Budget was not passed to seahorse.Compact,
it defaulted to `tokensBefore * 0.75`, making `tokensBefore > budget`
always true and causing condensed compaction to trigger unnecessarily.
Changes:
- context_seahorse.go: Forward Budget from CompactRequest to CompactInput
- loop.go: Pass Budget (ContextWindow) in all 3 Compact calls
- Add test verifying condensed is skipped when tokens < threshold
- Fix lint issues in store.go and store_test.go
* fix(seahorse): add mutex for assembler lazy initialization
Issue #5 from PR review: The check-then-create pattern for e.assembler
was a data race when multiple goroutines called Assemble() concurrently:
if e.assembler == nil {
e.assembler = &Assembler{...}
}
Changes:
- Add assemblerMu sync.Mutex to Engine struct
- Add initAssemblerOnce() using double-checked locking (same pattern as initCompactionOnce)
- Add TestAssemblerLazyInitRace to verify thread-safety
* fix(seahorse): handle non-consecutive depths in selectShallowestCondensationCandidate
Issue #8 from PR review: the loop iterated depth 0, 1, 2... assuming
consecutive keys, but break when key was missing caused deeper depths
to never be checked.
Fix: collect all existing depth keys, sort, then iterate in order.
* fix(seahorse): wrap DeleteMessagesAfterID and appendContextItems in transactions
- DeleteMessagesAfterID: wrap all DELETE operations in a transaction for
atomicity, remove redundant manual FTS delete (handled by trigger)
- appendContextItems: use transaction to fix read-then-write race condition
- Add GetMaxOrdinalTx and resolveItemTokenCountTx for transaction-scoped queries
- Remove unused resolveItemTokenCount function
Fixes PR review issues 6 and 7.
* fix(seahorse): derive readable content from Parts and cap CompactUntilUnder iterations
- Derive readable content from MessageParts in AddMessageWithParts so
FTS5 indexing and summary formatting can access tool call information
- formatMessagesForSummary and truncateSummary now fall back to Parts
when Content is empty, fixing blank summaries for Part-based messages
- Add MaxCompactIterations (20) to prevent CompactUntilUnder infinite
loops; exceeded iterations are logged as warnings
2026-04-05 01:05:16 +00:00
|
|
|
msgTokens += EstimateMessageTokens(m)
|
2026-03-13 06:20:24 +00:00
|
|
|
}
|
|
|
|
|
|
feat(seahorse): implement short-term memory engine (LCM) (#2285)
* feat(seahorse): implement short-term memory engine of seahorse
Add pkg/seahorse/ module implementing a SQLite-backed DAG-based summary
hierarchy for context management, ported from lossless-claw's LCM design:
- types.go + short_constants.go: core types (Message, Summary, Conversation,
ContextItem) and configuration constants (fanout, token targets, thresholds)
- migration.go: idempotent DB schema with FTS5 trigram tokenizer for CJK
- store.go: full SQLite CRUD (conversations, messages, summaries DAG,
context_items with ordinal gap numbering, FTS5 search)
- short_engine.go: Engine lifecycle (NewEngine, Ingest, Assemble, Compact),
session pattern filtering (ignore/stateless glob→regex compilation),
per-session mutex via sync.Map
- short_assembler.go: budget-aware context assembly with fresh tail protection
(32 messages), oldest-first eviction, summary XML formatting, RebuildContextItems
- short_compaction.go: leaf compaction (messages→summary) and condensed
compaction (summaries→higher-level summary), 3-level LLM escalation,
CompactUntilUnder for emergency overflow
- short_retrieval.go: lookupByID, FTS5/LIKE search, recursive expand with
token cap
- context_seahorse.go: agent.ContextManager adapter, registered as "seahorse",
provider↔seahorse message type conversion (ToolCalls, tool_result)
* fix(seahorse): correct 3 adapter bugs in context management
- TokenCount: use full message (Content+ToolCalls+Media) instead of Content-only
- Empty Content: rebuild Content from tool_result Parts when stored empty
- Duplicate summaries: summaries only in Summary field, not in History messages
- Grep: fix SearchResult.Snippet→Content for summaries
- Schema: fix FTS5 SQL uses VIRTUAL TABLE not TEMP TABLE
- TestFTS5SQLConstants: verify FTS5 SQL syntax correctness
- Test: fix flaky TestCompactLeaf
* fix(agent): ingest steering messages into seahorse SQLite
Steering messages were only persisted to session JSONL but not ingested
into seahorse SQLite, causing them to be missing from context assembly.
Added `ts.ingestMessage(turnCtx, al, pm)` call in the steering message
injection block alongside the existing JSONL persistence.
Test: TestSeahorseSteeringMessageIngested verifies steering messages
appear in seahorse SQLite DB after being processed.
* fix(seahorse): address 3 blocking bugs from code review
- Fix resequenceContextItemsTx scan error handling (store.go:850)
Changed `return err` to `return scanErr` to properly propagate scan errors
instead of returning nil (which silently corrupts data)
- Fix sql.NullString for INTEGER column (store.go:847)
Changed `mid` from sql.NullString to sql.NullInt64 since message_id
is INTEGER in schema. Removed unnecessary strconv.ParseInt call.
- Fix compactCondensed fallback deleting non-candidate items
Added ReplaceContextItemsWithSummary method for per-item deletion
when candidates are not contiguous in ordinal space.
Optimized to use range deletion when candidates are consecutive.
* fix(seahorse): pass Budget to Compact for correct condensed threshold
Issue #4 from PR review: When Budget was not passed to seahorse.Compact,
it defaulted to `tokensBefore * 0.75`, making `tokensBefore > budget`
always true and causing condensed compaction to trigger unnecessarily.
Changes:
- context_seahorse.go: Forward Budget from CompactRequest to CompactInput
- loop.go: Pass Budget (ContextWindow) in all 3 Compact calls
- Add test verifying condensed is skipped when tokens < threshold
- Fix lint issues in store.go and store_test.go
* fix(seahorse): add mutex for assembler lazy initialization
Issue #5 from PR review: The check-then-create pattern for e.assembler
was a data race when multiple goroutines called Assemble() concurrently:
if e.assembler == nil {
e.assembler = &Assembler{...}
}
Changes:
- Add assemblerMu sync.Mutex to Engine struct
- Add initAssemblerOnce() using double-checked locking (same pattern as initCompactionOnce)
- Add TestAssemblerLazyInitRace to verify thread-safety
* fix(seahorse): handle non-consecutive depths in selectShallowestCondensationCandidate
Issue #8 from PR review: the loop iterated depth 0, 1, 2... assuming
consecutive keys, but break when key was missing caused deeper depths
to never be checked.
Fix: collect all existing depth keys, sort, then iterate in order.
* fix(seahorse): wrap DeleteMessagesAfterID and appendContextItems in transactions
- DeleteMessagesAfterID: wrap all DELETE operations in a transaction for
atomicity, remove redundant manual FTS delete (handled by trigger)
- appendContextItems: use transaction to fix read-then-write race condition
- Add GetMaxOrdinalTx and resolveItemTokenCountTx for transaction-scoped queries
- Remove unused resolveItemTokenCount function
Fixes PR review issues 6 and 7.
* fix(seahorse): derive readable content from Parts and cap CompactUntilUnder iterations
- Derive readable content from MessageParts in AddMessageWithParts so
FTS5 indexing and summary formatting can access tool call information
- formatMessagesForSummary and truncateSummary now fall back to Parts
when Content is empty, fixing blank summaries for Part-based messages
- Add MaxCompactIterations (20) to prevent CompactUntilUnder infinite
loops; exceeded iterations are logged as warnings
2026-04-05 01:05:16 +00:00
|
|
|
toolTokens := EstimateToolDefsTokens(toolDefs)
|
2026-03-13 06:20:24 +00:00
|
|
|
total := msgTokens + toolTokens + maxTokens
|
|
|
|
|
|
|
|
|
|
return total > contextWindow
|
|
|
|
|
}
|
2026-05-18 19:11:21 +00:00
|
|
|
|
|
|
|
|
// trimHistoryToFitContextWindow rebuilds the prompt from progressively newer
|
|
|
|
|
// history slices until it fits within the context window. Oldest complete turns
|
|
|
|
|
// are dropped first so tool-call sequences remain intact.
|
|
|
|
|
func trimHistoryToFitContextWindow(
|
|
|
|
|
history []providers.Message,
|
|
|
|
|
build func([]providers.Message) []providers.Message,
|
|
|
|
|
contextWindow int,
|
|
|
|
|
toolDefs []providers.ToolDefinition,
|
|
|
|
|
maxTokens int,
|
|
|
|
|
) ([]providers.Message, []providers.Message, bool) {
|
|
|
|
|
messages := build(history)
|
|
|
|
|
if !isOverContextBudget(contextWindow, messages, toolDefs, maxTokens) {
|
|
|
|
|
return history, messages, true
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
trimmedHistory := append([]providers.Message(nil), history...)
|
|
|
|
|
for len(trimmedHistory) > 0 {
|
|
|
|
|
dropUntil := nextHistoryTrimStart(trimmedHistory)
|
|
|
|
|
if dropUntil <= 0 || dropUntil >= len(trimmedHistory) {
|
|
|
|
|
trimmedHistory = nil
|
|
|
|
|
} else {
|
|
|
|
|
trimmedHistory = append([]providers.Message(nil), trimmedHistory[dropUntil:]...)
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
messages = build(trimmedHistory)
|
|
|
|
|
if !isOverContextBudget(contextWindow, messages, toolDefs, maxTokens) {
|
|
|
|
|
return trimmedHistory, messages, true
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return nil, messages, false
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
func nextHistoryTrimStart(history []providers.Message) int {
|
|
|
|
|
if len(history) == 0 {
|
|
|
|
|
return 0
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
turns := parseTurnBoundaries(history)
|
|
|
|
|
if len(turns) >= 2 {
|
|
|
|
|
return turns[1]
|
|
|
|
|
}
|
|
|
|
|
if len(turns) == 1 {
|
|
|
|
|
if turns[0] > 0 {
|
|
|
|
|
return turns[0]
|
|
|
|
|
}
|
|
|
|
|
return len(history)
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return len(history)
|
|
|
|
|
}
|