fix(context): expose history tokens and remove leaked state files

Address remaining review feedback: 1) Add HistoryTokens field to ContextUsage/ContextStats, showing history-only token count in /context and frontend UI alongside SummarizeAtTokens so users can see the actual summarization trigger comparison. 2) Remove .codebuddy/github-contribute/ state files accidentally included in the PR.

.codebuddy/github-contribute/config.md

@@ -1,17 +0,0 @@
-# GitHub Contribution Config — sipeed/picoclaw
-
-- run_mode: resume
-- credential_source: session_override_env
-- github_username: chengzhichao-xydt
-- github_token: [session-provided]
-- auth_mode: env_pat
-- pat_type: classic
-- max_open_prs_per_project: 5
-- max_prs_waiting_for_first_review: 3
-- tasks_per_batch: 3
-- pr_recheck_interval_minutes: 20
-- report_language: zh-CN
-- report_timezone: Asia/Shanghai (UTC+8)
-- default_branch_prefix: codex
-- default_risk_level: low
-- github_communication_language: en

.codebuddy/github-contribute/handoff.md

@@ -1,12 +0,0 @@
-# Handoff — sipeed/picoclaw
-
-Last updated: 2026-06-05 09:30 UTC+8
-Run mode: resume (fix round complete)
-
-## Quick Resume
-- Repo: sipeed/picoclaw (fork: chengzhichao-xydt/picoclaw)
-- Upstream main: 5224b9a4
-- Open PRs: #3001, #2985 — both pushed with review fixes
-- Both waiting for afjcjsbx re-review
-- 7 merged PRs this round
-- Next: monitor CI/review, scan new candidates if desired

.codebuddy/github-contribute/journal.md

@@ -1,22 +0,0 @@
-# Journal — sipeed/picoclaw
-
-## 2026-06-05 08:58 UTC+8 — Resume + Fix round
-
-- run_mode=resume, 2 open PRs identified
-- Initialized repo-local state directory (.codebuddy/github-contribute)
-- Cleaned up merged local branches
-
-## 2026-06-05 09:05 UTC+8 — #3001 fix
-- Added TestShellTool_SchemelessURLDetection to shell_test.go
-- Covers all 7 web schemes + multi-URL commands
-- Test PASS, pushed a6735517, commented on PR
-
-## 2026-06-05 09:25 UTC+8 — #2985 fix
-- Issue 1: Added comment explaining summarizeAtTokens vs UsedTokens difference
-- Issue 2: Replaced hardcoded strings with i18n keys (5 locale files)
-- Issue 3: Added SummarizeAtTokens field to pico_test.go with assertion
-- Build PASS, pico test PASS, pushed 296a8ae2, commented on PR
-
-## 2026-06-05 09:30 UTC+8 — State update
-- Both PRs now in 'wait' — awaiting afjcjsbx re-review
-- Ready for next batch of candidate scanning

.codebuddy/github-contribute/pr-board.md

@@ -1,15 +0,0 @@
-# PR Board — sipeed/picoclaw
-
-## awaiting_maintainer_review (2)
-
-### #3001 — fix(tools): allow scheme-less URLs in workspace guard
-- Branch: codex/workspace-guard-schemeless-url
-- Fix: Added TestShellTool_SchemelessURLDetection (7 schemes + multi-URL)
-- Pushed: a6735517 | Comment: https://github.com/sipeed/picoclaw/pull/3001#issuecomment-4627308317
-- Next: wait for afjcjsbx re-review
-
-### #2985 — fix(context): show both summarize and compress thresholds in /context
-- Branch: codex/context-show-summarize-threshold
-- Fix: ① Threshold alignment comment ② i18n (5 locales) ③ Test coverage for summarize_at_tokens
-- Pushed: 296a8ae2 | Comment: https://github.com/sipeed/picoclaw/pull/2985#issuecomment-4627346564
-- Next: wait for afjcjsbx re-review

.codebuddy/github-contribute/queue.md

@@ -1,16 +0,0 @@
-# Active Queue — sipeed/picoclaw
-
-## Active (0)
-
-## Waiting for maintainer review (2)
-
-| ID | PR | Title | Status | Action |
-|----|-----|-------|--------|--------|
-| T1 | #3001 | fix(tools): allow scheme-less URLs in workspace guard | ✅ fix pushed (unit tests added) | wait |
-| T2 | #2985 | fix(context): show both summarize and compress thresholds | ✅ fix pushed (3 review issues) | wait |
-
-## Merged (7)
-#3000, #2999, #2996, #2995, #2992, #2991, #2986
-
-## Blocked (0)
-None

pkg/agent/agent_command.go

@@ -352,6 +352,7 @@ func (al *AgentLoop) buildCommandsRuntime(
 			return &commands.ContextStats{
 				UsedTokens:        usage.UsedTokens,
 				TotalTokens:       usage.TotalTokens,
+				HistoryTokens:     usage.HistoryTokens,
 				CompressAtTokens:  usage.CompressAtTokens,
 				SummarizeAtTokens: usage.SummarizeAtTokens,
 				UsedPercent:       usage.UsedPercent,

pkg/agent/context_usage.go

@@ -64,12 +64,9 @@ func computeContextUsage(agent *AgentInstance, sessionKey string) *bus.ContextUs
 	// summarizeAt = soft summarization trigger: matches maybeSummarize's
 	// threshold (contextWindow * SummarizeTokenPercent / 100).
 	//
-	// NOTE: The engine's maybeSummarize compares this threshold against
+	// The engine compares this against history-message tokens ONLY (not
-	// history-message tokens only (via estimateTokens), while UsedTokens
+	// UsedTokens).  HistoryTokens is exposed alongside UsedTokens so the
-	// (shown in /context) includes system prompt, summary, and tool
+	// UI can show both values and avoid user confusion.
-	// definitions on top of history tokens.  A "Used > SummarizeAt" display
-	// does not necessarily mean summarization will fire — the engine may
-	// still consider the history-token budget to be under threshold.
 	summarizeAt := contextWindow * agent.SummarizeTokenPercent / 100
 	if summarizeAt <= 0 {
 		summarizeAt = compressAt
@@ -86,6 +83,7 @@ func computeContextUsage(agent *AgentInstance, sessionKey string) *bus.ContextUs
 	return &bus.ContextUsage{
 		UsedTokens:        usedTokens,
 		TotalTokens:       contextWindow,
+		HistoryTokens:     historyTokens,
 		CompressAtTokens:  compressAt,
 		SummarizeAtTokens: summarizeAt,
 		UsedPercent:       usedPercent,

pkg/bus/types.go

@@ -66,8 +66,9 @@ type OutboundScope struct {
 type ContextUsage struct {
 	UsedTokens        int `json:"used_tokens"`
 	TotalTokens       int `json:"total_tokens"`        // model context window
+	HistoryTokens     int `json:"history_tokens"`      // history-message tokens only (what maybeSummarize checks)
 	CompressAtTokens  int `json:"compress_at_tokens"`  // hard budget compression threshold (contextWindow - maxTokens)
-	SummarizeAtTokens int `json:"summarize_at_tokens"` // soft summarization trigger (contextWindow * summarizeTokenPercent / 100)
+	SummarizeAtTokens int `json:"summarize_at_tokens"` // soft summarization trigger (vs history tokens)
 	UsedPercent       int `json:"used_percent"`        // 0-100, relative to compressAt
 }
 

pkg/channels/pico/pico.go

@@ -1396,6 +1396,7 @@ func setContextUsagePayload(payload map[string]any, u *bus.ContextUsage) {
 	payload["context_usage"] = map[string]any{
 		"used_tokens":         u.UsedTokens,
 		"total_tokens":        u.TotalTokens,
+		"history_tokens":      u.HistoryTokens,
 		"compress_at_tokens":  u.CompressAtTokens,
 		"summarize_at_tokens": u.SummarizeAtTokens,
 		"used_percent":        u.UsedPercent,

pkg/channels/pico/pico_test.go

@@ -604,6 +604,7 @@ func TestBeginStream_FinalizeIncludesContextUsage(t *testing.T) {
 	if err := contextStreamer.FinalizeWithContext(context.Background(), "final", &bus.ContextUsage{
 		UsedTokens:        10,
 		TotalTokens:       100,
+		HistoryTokens:     5,
 		CompressAtTokens:  80,
 		SummarizeAtTokens: 60,
 		UsedPercent:       10,
@@ -628,6 +629,9 @@ func TestBeginStream_FinalizeIncludesContextUsage(t *testing.T) {
 	if got := rawUsage["used_tokens"]; got != float64(10) {
 		t.Fatalf("used_tokens = %#v, want 10", got)
 	}
+	if got := rawUsage["history_tokens"]; got != float64(5) {
+		t.Fatalf("history_tokens = %#v, want 5", got)
+	}
 	if got := rawUsage["summarize_at_tokens"]; got != float64(60) {
 		t.Fatalf("summarize_at_tokens = %#v, want 60", got)
 	}

pkg/commands/cmd_context.go

@@ -30,11 +30,12 @@ func formatContextStats(s *ContextStats) string {
 	}
 	usedWindowPercent := s.UsedTokens * 100 / max(s.TotalTokens, 1)
 	msg := fmt.Sprintf(
-		"Context usage  \nMessages: %d  \nUsed: ~%d / %d tokens (%d%%)  \nCompress at: %d tokens  \nSummarize at: %d tokens  \nCompression progress: %d%%  \nRemaining: ~%d tokens",
+		"Context usage  \nMessages: %d  \nUsed: ~%d / %d tokens (%d%%)  \nHistory: ~%d tokens  \nCompress at: %d tokens  \nSummarize at: %d tokens  \nCompression progress: %d%%  \nRemaining: ~%d tokens",
 		s.MessageCount,
 		s.UsedTokens,
 		s.TotalTokens,
 		usedWindowPercent,
+		s.HistoryTokens,
 		s.CompressAtTokens,
 		s.SummarizeAtTokens,
 		s.UsedPercent,

pkg/commands/runtime.go

@@ -31,6 +31,7 @@ type MCPToolInfo struct {
 type ContextStats struct {
 	UsedTokens        int
 	TotalTokens       int // model context window
+	HistoryTokens     int // history-only tokens (what maybeSummarize checks)
 	CompressAtTokens  int // hard budget compression threshold
 	SummarizeAtTokens int // soft summarization trigger
 	UsedPercent       int // 0-100

web/frontend/src/components/chat/context-usage-ring.tsx

@@ -146,6 +146,14 @@ export function ContextUsageRing({
           </div>
 
           <div className="mt-2 space-y-0.5">
+            {usage.history_tokens != null && usage.history_tokens > 0 && (
+              <div className="flex items-center justify-between text-[10px]">
+                <span className="text-muted-foreground">History</span>
+                <span className="tabular-nums">
+                  {formatTokens(usage.history_tokens)}
+                </span>
+              </div>
+            )}
             <div className="flex items-center justify-between text-[10px]">
               <span className="text-muted-foreground">{t("chat.contextCompressAt")}</span>
               <span className="tabular-nums">

web/frontend/src/features/chat/protocol.ts

@@ -78,6 +78,7 @@ function parseContextUsage(
   return {
     used_tokens: used,
     total_tokens: total,
+    history_tokens: obj.history_tokens != null ? Number(obj.history_tokens) : undefined,
     compress_at_tokens: Number(obj.compress_at_tokens) || 0,
     summarize_at_tokens: obj.summarize_at_tokens != null ? Number(obj.summarize_at_tokens) : undefined,
     used_percent: Number(obj.used_percent) || 0,

web/frontend/src/store/chat.ts

@@ -52,6 +52,7 @@ export interface ChatMessage {
 export interface ContextUsage {
   used_tokens: number
   total_tokens: number
+  history_tokens?: number
   compress_at_tokens: number
   summarize_at_tokens?: number
   used_percent: number