@martian-engineering/lossless-claw 0.11.3 → 0.13.0

package/docs/architecture.md

@@ -212,6 +212,17 @@ LCM handles crash recovery through **bootstrap reconciliation**:
 
 This handles the case where OpenClaw wrote messages to the session file but crashed before LCM could persist them.
 
+For forked child sessions, LCM treats a host-copied parent JSONL branch as a
+first-time bootstrap source and imports only the newest messages that fit within
+`bootstrapMaxTokens`. That keeps child LCM state bounded even if the host fork
+payload still contains a long raw parent branch. The remaining fork-continuity
+contract belongs to the host: when lossless-claw advertises the
+`subagent-spawn` requirement for `thread-bootstrap-projection`, OpenClaw should
+bootstrap the child model thread from the context-engine projection rather than
+from the raw copied transcript. If the host cannot provide that capability,
+lossless-claw can preserve bounded durable state, but it cannot stop the host
+from replaying raw JSONL into the model before assembly.
+
 ## Operation serialization
 
 All mutating operations (ingest, compact) are serialized per-session using a promise queue. This prevents races between concurrent afterTurn/compact calls for the same conversation without blocking operations on different conversations.

package/docs/configuration.md

@@ -30,6 +30,23 @@ Most installations only need to override a handful of keys. If you want a comple
   "statelessSessionPatterns": [],
   "skipStatelessSessions": true,
   "contextThreshold": 0.75,
+  "contextThresholdOverrides": [
+    {
+      "name": "large-context-models",
+      "match": { "modelContextWindowMin": 900000 },
+      "contextThreshold": 0.15
+    },
+    {
+      "name": "small-context-models",
+      "match": { "modelContextWindowMax": 250000 },
+      "contextThreshold": 0.2
+    },
+    {
+      "name": "telegram-sessions",
+      "match": { "sessionPattern": "agent:*:telegram:**" },
+      "contextThreshold": 0.3
+    }
+  ],
   "freshTailCount": 64,
   "freshTailMaxTokens": 24000,
   "promptAwareEviction": false,
@@ -58,14 +75,20 @@ Most installations only need to override a handful of keys. If you want a comple
   "expansionModel": "",
   "delegationTimeoutMs": 120000,
   "summaryTimeoutMs": 60000,
+  "summaryCallWindowMs": 600000,
+  "summaryMaxCallsPerWindow": 24,
+  "summarySpendBackoffMs": 1800000,
   "timezone": "America/Los_Angeles",
   "pruneHeartbeatOk": false,
   "transcriptGcEnabled": false,
+  "enableSummaryThinking": true,
   "maxAssemblyTokenBudget": 30000,
   "summaryMaxOverageFactor": 3,
   "customInstructions": "",
   "circuitBreakerThreshold": 5,
   "circuitBreakerCooldownMs": 1800000,
+  "replayFloodThresholdExternal": 3,
+  "replayFloodThresholdInternal": 32,
   "fallbackProviders": [],
   "proactiveThresholdCompactionMode": "deferred",
   "autoRotateSessionFiles": {
@@ -75,6 +98,11 @@ Most installations only need to override a handful of keys. If you want a comple
     "startup": "rotate",
     "runtime": "rotate"
   },
+  "independentLogFile": {
+    "enabled": true,
+    "file": "/tmp/openclaw/lossless-claw-2026-05-19.log",
+    "maxFileBytes": 104857600
+  },
   "cacheAwareCompaction": {
     "enabled": true,
     "cacheTTLSeconds": 300,
@@ -87,7 +115,13 @@ Most installations only need to override a handful of keys. If you want a comple
   "dynamicLeafChunkTokens": {
     "enabled": true,
     "max": 40000
-  }
+  },
+  "stripInjectedContextTags": [
+    "active_memory_plugin",
+    "relevant-memories",
+    "relevant_memories",
+    "hindsight_memories"
+  ]
 }
 ```
 
@@ -139,16 +173,22 @@ openclaw plugins install --link /path/to/lossless-claw
 | `timezone` | `string` | `TZ` or system timezone | `TZ` | IANA timezone used for timestamp rendering in summaries. |
 | `pruneHeartbeatOk` | `boolean` | `false` | `LCM_PRUNE_HEARTBEAT_OK` | Retroactively removes `HEARTBEAT_OK` turn cycles from persisted storage. |
 | `transcriptGcEnabled` | `boolean` | `false` | `LCM_TRANSCRIPT_GC_ENABLED` | Enables transcript rewrite GC during `maintain()`; disabled by default so transcript rewrites stay opt-in. |
+| `enableSummaryThinking` | `boolean` | `true` | `LCM_ENABLE_SUMMARY_THINKING` | When true, requests low reasoning budget from the model during summarization calls. Set to false to disable reasoning and keep summarization output concise. |
 | `proactiveThresholdCompactionMode` | `"deferred" \| "inline"` | `"deferred"` | `LCM_PROACTIVE_THRESHOLD_COMPACTION_MODE` | Controls whether proactive threshold compaction is deferred into maintenance debt by default or run inline for legacy behavior. |
 | `autoRotateSessionFiles.enabled` | `boolean` | `true` | `LCM_AUTO_ROTATE_SESSION_FILES_ENABLED` | Enables automatic rotation for oversized LCM-managed session JSONL files. |
 | `autoRotateSessionFiles.createBackups` | `boolean` | `false` | `LCM_AUTO_ROTATE_SESSION_FILES_CREATE_BACKUPS` | Creates or replaces the rolling `rotate-latest` SQLite backup before automatic session-file rotation. Manual `/lcm rotate` backups are always created. |
 | `autoRotateSessionFiles.sizeBytes` | `integer` | `2097152` | `LCM_AUTO_ROTATE_SESSION_FILES_SIZE_BYTES` | Byte threshold that triggers automatic session-file rotation. |
 | `autoRotateSessionFiles.startup` | `"rotate" \| "warn" \| "off"` | `"rotate"` | `LCM_AUTO_ROTATE_SESSION_FILES_STARTUP` | Startup behavior for oversized indexed OpenClaw session transcripts that also have active LCM bootstrap state. |
-| `autoRotateSessionFiles.runtime` | `"rotate" \| "warn" \| "off"` | `"rotate"` | `LCM_AUTO_ROTATE_SESSION_FILES_RUNTIME` | Runtime behavior after `afterTurn()` and `maintain()` check the current transcript size. |
+| `autoRotateSessionFiles.runtime` | `"rotate" \| "warn" \| "off"` | `"rotate"` | `LCM_AUTO_ROTATE_SESSION_FILES_RUNTIME` | Runtime behavior after post-turn checks. Runtime `rotate` logs deferral for active session JSONL rewrites and leaves direct rotation to startup or manual `/lcm rotate`. |
+| `independentLogFile.enabled` | `boolean` | `true` | `LCM_LOG_FILE_ENABLED` | Writes lossless-claw JSONL logs to an independent plugin-owned file in addition to OpenClaw's runtime logger. |
+| `independentLogFile.file` | `string` | `/tmp/openclaw/lossless-claw-YYYY-MM-DD.log` | `LCM_LOG_FILE` | Optional log path. A dated `lossless-claw-YYYY-MM-DD.log` path rolls over daily. |
+| `independentLogFile.maxFileBytes` | `integer` | `104857600` | `LCM_LOG_MAX_FILE_BYTES` | Size threshold for rotating the active lossless-claw log file to `.1.log` through `.5.log`. |
 
 > **Multi-profile note:** `OPENCLAW_STATE_DIR` (set by the host OpenClaw gateway) controls where state is stored. When two gateways run on the same host (e.g. separate bot personas), each gateway sets its own `OPENCLAW_STATE_DIR` and lossless-claw automatically uses that directory for the database, large-file payloads, auth-profile lookups, and legacy secrets — no per-profile plugin config is needed.
 
-Automatic session-file rotation rewrites only the live session transcript, keeps the active LCM conversation and durable history intact, and refreshes the bootstrap checkpoint. Startup rotation first scans OpenClaw's current indexed session stores for configured agents, then intersects those candidates with active LCM conversations and matching bootstrap file mappings. Automatic rotation does not create a SQLite backup by default; set `autoRotateSessionFiles.createBackups` to `true` to make runtime rotation replace the rolling `rotate-latest` backup and to make startup rotation create one pre-rotation LCM database backup for the batch before any transcript is rewritten. Manual `/lcm rotate` always keeps its backup-backed behavior regardless of this flag. Rotation never runs for ignored sessions, stateless sessions, or sessions without active LCM state. The preserved JSONL tail follows the existing rotate behavior, which is controlled by `freshTailCount`.
+Automatic session-file rotation rewrites only the live session transcript, keeps the active LCM conversation and durable history intact, and refreshes the bootstrap checkpoint. Before manual or startup rewrites, rotation forces leaf-only compaction for raw context outside the preserved tail so trimmed transcript messages are covered by LCM summaries without running unrelated summary-condensation passes. Startup rotation first scans OpenClaw's current indexed session stores for configured agents, then intersects those candidates with active LCM conversations and matching bootstrap file mappings. Runtime rotation checks from `afterTurn()` and `maintain()` intentionally do not directly rewrite active session JSONL because embedded prompt-lock fences can still be open while tool-call loops and host background maintenance overlap; runtime `rotate` logs a deferral until startup, manual `/lcm rotate`, or a future host-owned full-transcript rewrite primitive is available. Automatic rotation does not create a SQLite backup by default; set `autoRotateSessionFiles.createBackups` to `true` to make startup rotation create one pre-rotation LCM database backup for the batch before any transcript is rewritten. Manual `/lcm rotate` always keeps its backup-backed behavior regardless of this flag. Rotation never runs for ignored sessions, stateless sessions, or sessions without active LCM state. The preserved JSONL tail follows the existing rotate behavior, which is controlled by `freshTailCount`. Transcript GC uses the host-provided `rewriteTranscriptEntries` primitive and defers until host-approved background maintenance when `transcriptGcEnabled` is enabled.
+
+Lossless-claw writes routine operational JSONL logs by default at `/tmp/openclaw/lossless-claw-YYYY-MM-DD.log`, beside OpenClaw's `/tmp/openclaw/openclaw-YYYY-MM-DD.log`. Routine info and debug lines go to the independent file instead of the shared OpenClaw log. Startup banners and warning/error lines still go through OpenClaw's runtime logger so gateway-level startup and failure diagnostics remain visible. The independent file follows the same practical rotation model as OpenClaw: a dated filename rolls over when the local date changes, stale dated files are pruned after 3 days, and an oversized active file is rotated through `.1.log` to `.5.log`.
 
 Every automatic decision emits grep-able log lines prefixed with `[lcm] auto-rotate:`. Startup emits one compact summary line with `phase=startup`, `action=summary`, `scanned`, `eligible`, `rotated`, `warned`, `skipped`, `durationMs`, `bytesRemoved`, and backup fields when a batch backup was created; quiet skips such as missing files, missing bootstrap mappings, and below-threshold files are counted there instead of producing one line per candidate. Rotation detail lines include `phase`, `action`, `sessionId`, `sessionKey`, `sessionFile`, `sizeBytes`, `thresholdBytes`, `durationMs`, `backupPath`, `bytesRemoved`, `preservedTailMessageCount`, and `checkpointSize`; real warning lines include the same available context plus `reason` or `error`.
 
@@ -157,6 +197,7 @@ Every automatic decision emits grep-able log lines prefixed with `[lcm] auto-rot
 | Key | Type | Default | Env override | Purpose |
 | --- | --- | --- | --- | --- |
 | `contextThreshold` | `number` | `0.75` | `LCM_CONTEXT_THRESHOLD` | Fraction of the active model context window that triggers compaction. |
+| `contextThresholdOverrides` | `Array<{ name?: string; match: object; contextThreshold: number }>` | `[]` | none | Optional ordered rules that override `contextThreshold` by model id, model context-window range, or session glob pattern. |
 | `freshTailCount` | `integer` | `64` | `LCM_FRESH_TAIL_COUNT` | Number of newest messages always kept raw. |
 | `freshTailMaxTokens` | `integer` | unset | `LCM_FRESH_TAIL_MAX_TOKENS` | Optional token cap for the protected fresh tail. The newest message is always preserved even if it exceeds the cap. |
 | `promptAwareEviction` | `boolean` | `false` | `LCM_PROMPT_AWARE_EVICTION_ENABLED` | When enabled, budget-constrained assembly keeps older evictable items by prompt relevance instead of pure chronology. This improves retrieval under tight budgets, but it can reduce prompt-cache hit rates because the preserved prefix changes as prompts change. |
@@ -180,6 +221,13 @@ Every automatic decision emits grep-able log lines prefixed with `[lcm] auto-rot
 | `maxAssemblyTokenBudget` | `integer` | unset | `LCM_MAX_ASSEMBLY_TOKEN_BUDGET` | Optional hard cap for assembly and threshold evaluation, useful with smaller-context models. |
 | `maxExpandTokens` | `integer` | `4000` | `LCM_MAX_EXPAND_TOKENS` | Default token cap for `lcm_expand_query` responses. |
 
+Forked child transcripts are also bounded by `bootstrapMaxTokens` when a host
+copies a raw parent JSONL branch into the child file. This protects the LCM
+database from importing unbounded parent history, but the host must still honor
+the `thread-bootstrap-projection` context-engine capability for subagent or
+thread forks so the model starts from the LCM-assembled compact view instead of
+the raw copied transcript.
+
 ### Model selection, execution, and prompts
 
 | Key | Type | Default | Env override | Purpose |
@@ -192,6 +240,9 @@ Every automatic decision emits grep-able log lines prefixed with `[lcm] auto-rot
 | `expansionProvider` | `string` | `""` | `LCM_EXPANSION_PROVIDER` | `lcm_expand_query` sub-agent provider hint for bare model names. |
 | `delegationTimeoutMs` | `integer` | `120000` | `LCM_DELEGATION_TIMEOUT_MS` | Maximum time to wait for delegated expansion work. `lcm_expand_query` advertises a dynamic tool `timeoutMs` default with 30 seconds of extra RPC headroom so OpenClaw's tool watchdog does not fire before this wait completes. |
 | `summaryTimeoutMs` | `integer` | `60000` | `LCM_SUMMARY_TIMEOUT_MS` | Maximum time to wait for one model-backed summarizer call. |
+| `summaryCallWindowMs` | `integer` | `600000` | `LCM_SUMMARY_CALL_WINDOW_MS` | Rolling window for the per-session summarization spend guard. |
+| `summaryMaxCallsPerWindow` | `integer` | `24` | `LCM_SUMMARY_MAX_CALLS_PER_WINDOW` | Maximum model-backed summarization calls per session/window before Lossless opens a non-auth spend backoff. |
+| `summarySpendBackoffMs` | `integer` | `1800000` | `LCM_SUMMARY_SPEND_BACKOFF_MS` | Cooldown after the summarization spend guard opens. |
 | `customInstructions` | `string` | `""` | `LCM_CUSTOM_INSTRUCTIONS` | Extra natural-language instructions injected into every summarization prompt. |
 
 Summary calls are executed through OpenClaw's `api.runtime.llm.complete` capability. If you configure an explicit Lossless summary model (`summaryModel`, `largeFileSummaryModel`, or `fallbackProviders`), OpenClaw must allow that runtime LLM override under `plugins.entries.lossless-claw.llm.allowModelOverride` and `plugins.entries.lossless-claw.llm.allowedModels`. `openclaw doctor --fix` can add the minimal policy entries for configured Lossless summary models. Delegated expansion calls use OpenClaw's runtime sub-agent layer; explicit `expansionModel` values require `plugins.entries.lossless-claw.subagent.allowModelOverride` and a matching `subagent.allowedModels` entry, or `"*"` if you intentionally trust any expansion target. `openclaw doctor --fix` can add the minimal subagent policy, and `lcm_expand_query` retries once without the override if the host rejects it.
@@ -203,6 +254,9 @@ Summary calls are executed through OpenClaw's `api.runtime.llm.complete` capabil
 | `fallbackProviders` | `Array<{ provider: string; model: string }>` | `[]` | `LCM_FALLBACK_PROVIDERS` | Explicit provider/model fallback chain for compaction summarization. Format for env vars is `provider/model,provider/model`. |
 | `circuitBreakerThreshold` | `integer` | `5` | `LCM_CIRCUIT_BREAKER_THRESHOLD` | Consecutive auth failures before the summarization circuit breaker trips. |
 | `circuitBreakerCooldownMs` | `integer` | `1800000` | `LCM_CIRCUIT_BREAKER_COOLDOWN_MS` | Cooldown before the summarization circuit breaker resets automatically. |
+| `stripInjectedContextTags` | `string[]` | `["active_memory_plugin", "relevant-memories", "relevant_memories", "hindsight_memories"]` | `LCM_STRIP_INJECTED_CONTEXT_TAGS` | XML tag names whose blocks are stripped from message content before compaction summarization. Memory/context plugins inject these via `prependContext`; stripping prevents ephemeral retrieval context from polluting compacted summaries. Env var format is comma-separated tag names. Set to `[]` (or empty env string) to disable. |
+| `replayFloodThresholdExternal` | `integer` | `3` | `LCM_REPLAY_FLOOD_THRESHOLD_EXTERNAL` | Max replay-like messages allowed in a single SQLite-second for `role=user` before `assertNoReplayTimestampFlood` refuses the batch. Defaults to `3` to preserve replay defense for third-partyly-rebroadcastable input. |
+| `replayFloodThresholdInternal` | `integer` | `32` | `LCM_REPLAY_FLOOD_THRESHOLD_INTERNAL` | Max identical messages allowed in a single SQLite-second for `role=tool/assistant/system` before the anti-replay guard refuses the batch. Defaults to `32` to absorb legitimate idempotent sub-agent bursts (same-second tool returns like `{"status":"ok"}`). |
 
 ### Nested objects
 
@@ -229,13 +283,16 @@ Summary calls are executed through OpenClaw's `api.runtime.llm.complete` capabil
 
 Automatic compaction is threshold-only:
 
-- `afterTurn()` evaluates `contextThreshold` against the active token budget
+- `afterTurn()` evaluates the resolved context threshold against the active token budget
 - below threshold, no automatic compaction runs and no leaf debt is recorded
 - at or above threshold, inline mode runs a threshold full sweep immediately
-- deferred mode records one coalesced `"threshold"` maintenance row and drains it in the background, `maintain()`, or pre-assembly
+- deferred mode records one coalesced `"threshold"` maintenance row and normally drains it in the background or host-approved `maintain()`
+- pre-assembly drain is reserved as an emergency safeguard when the live prompt is already over the active token budget
 
 Lossless still records prompt-cache telemetry for status and diagnostics, but cache hotness no longer delays threshold debt. Legacy `cacheAwareCompaction.*` and `dynamicLeafChunkTokens.*` settings remain accepted so existing OpenClaw config continues to load, but they do not change automatic compaction behavior.
 
+`contextThresholdOverrides` are optional and never replace the global fallback. Each rule's `match` object can include `model`, `modelContextWindowMin`, `modelContextWindowMax`, and `sessionPattern`; all fields in a rule must match. If several rules match, Lossless picks the highest-specificity rule, then the earliest rule in the array for ties. Exact `model` matches have higher specificity than `sessionPattern` matches, and session-pattern matches have higher specificity than context-window range matches. Threshold selection logs include the chosen threshold, source, rule index/name, token budget, threshold tokens, model, context-window value, and match reason.
+
 Full sweeps first run leaf passes until there are no more eligible raw-message chunks outside the fresh tail. Condensation is then driven by summarized-prefix pressure: the routine condensation phase obeys `sweepMaxDepth`, and if the summarized prefix still exceeds `summaryPrefixTargetTokens`, a pressure phase may use `condensedMinFanoutHard` and condense deeper. Total context pressure starts the sweep, but does not by itself force deeper condensation once the raw prefix has been summarized.
 
 A single sweep is bounded by both `maxSweepIterations` (a hard cap on summarizer passes) and `sweepDeadlineMs` (a wall-clock budget). When either limit is reached the sweep stops before starting another pass and returns the consistent partial result built so far, logging a `compactFullSweep stopped at …` warning. This keeps a slow or rate-limited summarizer from hanging the agent turn — remaining context pressure is picked up by the next sweep.
@@ -287,6 +344,8 @@ LCM_EXPANSION_MODEL=openai/gpt-5.4-mini
 - `*` matches any characters except `:`
 - `**` matches anything, including `:`
 
+Cron scheduler keys (`agent:<agent>:cron:<job>...`) are isolated automatically when a new runtime `sessionId` reuses the same `sessionKey`. Configure `ignoreSessionPatterns` for cron only when the run should bypass LCM entirely; leave cron sessions included when they need in-run compaction.
+
 Example:
 
 ```json
@@ -318,7 +377,8 @@ Lossless-claw now defaults `proactiveThresholdCompactionMode` to `deferred`.
 - deferred mode records a single coalesced maintenance debt row per conversation
 - new deferred compaction debt is only created for `contextThreshold` pressure and uses reason `"threshold"`
 - `maintain()` consumes threshold debt when the host explicitly opts in to deferred execution
-- `assemble()` consumes pending threshold debt before building the next prompt
+- `assemble()` leaves pending threshold debt for after-turn background drain or host-approved `maintain()` while the live prompt is still within budget
+- `assemble()` only consumes pending threshold debt synchronously as an emergency safeguard when the live prompt estimate is already over the active token budget
 - old non-threshold debt from earlier builds is revalidated; if the conversation is no longer over threshold, it is cleared as a no-op
 - `/lcm status` / `/lossless status` shows the current maintenance state, including pending/running/last-failure details
 - status output also surfaces the latest API/cache telemetry as diagnostics, not as a deferral gate

package/openclaw.plugin.json

@@ -1,5 +1,6 @@
 {
   "id": "lossless-claw",
+  "name": "Lossless Context Management",
   "kind": "context-engine",
   "activation": {
     "onStartup": true
@@ -24,6 +25,10 @@
       "label": "Context Threshold",
       "help": "Fraction of context window that triggers compaction (0.0–1.0)"
     },
+    "contextThresholdOverrides": {
+      "label": "Context Threshold Overrides",
+      "help": "Optional ordered rules that override contextThreshold by model, model context-window range, or session glob pattern"
+    },
     "sweepMaxDepth": {
       "label": "Sweep Max Depth",
       "help": "Preferred maximum condensation source depth during routine full sweeps (0 = leaf only, -1 = unlimited). Pressure sweeps may go deeper when summarized context remains above target."
@@ -168,6 +173,18 @@
       "label": "Summary Timeout (ms)",
       "help": "Maximum time to wait for a single model-backed LCM summarizer call before timing out"
     },
+    "summaryCallWindowMs": {
+      "label": "Summary Call Window (ms)",
+      "help": "Rolling window used by the summarization spend guard"
+    },
+    "summaryMaxCallsPerWindow": {
+      "label": "Summary Max Calls Per Window",
+      "help": "Maximum model-backed summarization calls per session/window before Lossless opens a non-auth spend backoff"
+    },
+    "summarySpendBackoffMs": {
+      "label": "Summary Spend Backoff (ms)",
+      "help": "Cooldown after the summarization spend guard opens"
+    },
     "maxAssemblyTokenBudget": {
       "label": "Max Assembly Token Budget",
       "help": "Hard ceiling for assembly token budget — caps runtime-provided and fallback budgets. Set for smaller context-window models (e.g., 30000 for 32k models)"
@@ -188,6 +205,14 @@
       "label": "Circuit Breaker Cooldown (ms)",
       "help": "Cooldown before the summarization circuit breaker auto-resets"
     },
+    "replayFloodThresholdExternal": {
+      "label": "Replay Flood Threshold (External)",
+      "help": "Max replay-like messages allowed in a single SQLite-second for third-party roles (role=user) before the anti-replay guard refuses the batch. Defaults to 3 to preserve replay defense for third-partyly-rebroadcastable input."
+    },
+    "replayFloodThresholdInternal": {
+      "label": "Replay Flood Threshold (Internal)",
+      "help": "Max identical messages allowed in a single SQLite-second for internal roles (role=tool/assistant/system) before the anti-replay guard refuses the batch. Defaults to 32 to absorb legitimate idempotent sub-agent bursts."
+    },
     "cacheAwareCompaction.enabled": {
       "label": "Cache-Aware Compaction (Deprecated)",
       "help": "Deprecated compatibility setting. Automatic compaction is now threshold-only and does not use prompt-cache hot/cold state."
@@ -236,6 +261,10 @@
       "label": "Transcript GC",
       "help": "Enable transcript rewrite GC during maintain(); disabled by default"
     },
+    "enableSummaryThinking": {
+      "label": "Enable Summary Thinking",
+      "help": "Request low reasoning budget from the model during summarization calls"
+    },
     "proactiveThresholdCompactionMode": {
       "label": "Proactive Threshold Compaction Mode",
       "help": "Choose deferred compaction debt by default or keep legacy inline proactive compaction"
@@ -260,9 +289,25 @@
       "label": "Runtime Auto-Rotate",
       "help": "Runtime behavior for oversized current LCM session files: rotate, warn, or off"
     },
+    "independentLogFile.enabled": {
+      "label": "Independent Log File",
+      "help": "Write lossless-claw JSONL logs to a plugin-owned file in addition to OpenClaw's runtime logger"
+    },
+    "independentLogFile.file": {
+      "label": "Independent Log Path",
+      "help": "Optional lossless-claw log path; defaults to /tmp/openclaw/lossless-claw-YYYY-MM-DD.log with daily rollover"
+    },
+    "independentLogFile.maxFileBytes": {
+      "label": "Independent Log Max Bytes",
+      "help": "Byte threshold for size rotation of the current lossless-claw log file (default: 104857600)"
+    },
     "fallbackProviders": {
       "label": "Fallback Providers",
       "help": "Explicit runtime LLM fallback provider/model pairs for compaction summarization; entries require plugins.entries.lossless-claw.llm policy"
+    },
+    "stripInjectedContextTags": {
+      "label": "Strip Injected Context Tags",
+      "help": "XML tag names whose blocks are stripped from messages before summarization. Covers memory/context plugin prepended blocks (active-memory, memory-lancedb, hindsight-openclaw). Set to [] to disable."
     }
   },
   "configSchema": {
@@ -277,6 +322,50 @@
         "minimum": 0,
         "maximum": 1
       },
+      "contextThresholdOverrides": {
+        "type": "array",
+        "items": {
+          "type": "object",
+          "additionalProperties": false,
+          "required": [
+            "match",
+            "contextThreshold"
+          ],
+          "properties": {
+            "name": {
+              "type": "string"
+            },
+            "match": {
+              "type": "object",
+              "additionalProperties": false,
+              "minProperties": 1,
+              "properties": {
+                "model": {
+                  "type": "string",
+                  "minLength": 1
+                },
+                "modelContextWindowMin": {
+                  "type": "integer",
+                  "minimum": 1
+                },
+                "modelContextWindowMax": {
+                  "type": "integer",
+                  "minimum": 1
+                },
+                "sessionPattern": {
+                  "type": "string",
+                  "minLength": 1
+                }
+              }
+            },
+            "contextThreshold": {
+              "type": "number",
+              "minimum": 0,
+              "maximum": 1
+            }
+          }
+        }
+      },
       "sweepMaxDepth": {
         "type": "integer",
         "minimum": -1
@@ -406,6 +495,18 @@
         "type": "integer",
         "minimum": 1
       },
+      "summaryCallWindowMs": {
+        "type": "integer",
+        "minimum": 1
+      },
+      "summaryMaxCallsPerWindow": {
+        "type": "integer",
+        "minimum": 1
+      },
+      "summarySpendBackoffMs": {
+        "type": "integer",
+        "minimum": 1
+      },
       "maxAssemblyTokenBudget": {
         "type": "integer",
         "minimum": 1000
@@ -425,6 +526,14 @@
         "type": "integer",
         "minimum": 1
       },
+      "replayFloodThresholdExternal": {
+        "type": "integer",
+        "minimum": 1
+      },
+      "replayFloodThresholdInternal": {
+        "type": "integer",
+        "minimum": 1
+      },
       "cacheAwareCompaction": {
         "type": "object",
         "additionalProperties": false,
@@ -482,6 +591,9 @@
       "transcriptGcEnabled": {
         "type": "boolean"
       },
+      "enableSummaryThinking": {
+        "type": "boolean"
+      },
       "proactiveThresholdCompactionMode": {
         "type": "string",
         "enum": [
@@ -521,6 +633,22 @@
           }
         }
       },
+      "independentLogFile": {
+        "type": "object",
+        "additionalProperties": false,
+        "properties": {
+          "enabled": {
+            "type": "boolean"
+          },
+          "file": {
+            "type": "string"
+          },
+          "maxFileBytes": {
+            "type": "integer",
+            "minimum": 1
+          }
+        }
+      },
       "databasePath": {
         "description": "Path to LCM SQLite database (preferred key; alias of dbPath, default: <OPENCLAW_STATE_DIR>/lcm.db)",
         "type": "string"
@@ -540,6 +668,13 @@
           "required": ["provider", "model"],
           "additionalProperties": false
         }
+      },
+      "stripInjectedContextTags": {
+        "description": "XML tag names whose blocks are stripped from messages before summarization. Memory/context plugins prepend tagged blocks via prependContext that should not leak into compacted summaries.",
+        "type": "array",
+        "items": {
+          "type": "string"
+        }
       }
     }
   }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@martian-engineering/lossless-claw",
-  "version": "0.11.3",
+  "version": "0.13.0",
   "description": "Lossless Context Management plugin for OpenClaw — DAG-based conversation summarization with threshold compaction",
   "type": "module",
   "main": "dist/index.js",
@@ -18,8 +18,10 @@
   "scripts": {
     "build": "esbuild index.ts --bundle --platform=node --target=node22 --format=esm --outfile=dist/index.js --external:openclaw --external:\"@earendil-works/*\" --minify-whitespace",
     "changeset": "changeset",
-    "release:verify": "npm run build && npm test && npm pack --dry-run",
+    "plugin-inspector:ci": "npm exec --yes --package @openclaw/plugin-inspector@0.3.11 -- plugin-inspector ci --plugin-root . --out plugin-inspector-reports",
+    "release:verify": "npm run typecheck && npm run build && npm test && npm pack --dry-run",
     "test": "vitest run --dir test",
+    "typecheck": "tsc --noEmit --pretty false",
     "version-packages": "changeset version"
   },
   "files": [
@@ -33,20 +35,20 @@
     "LICENSE"
   ],
   "dependencies": {
-    "@earendil-works/pi-agent-core": ">=0.74 <1",
-    "@earendil-works/pi-ai": ">=0.74 <1",
-    "@earendil-works/pi-coding-agent": ">=0.74 <1",
     "@sinclair/typebox": "0.34.48"
   },
   "devDependencies": {
     "@changesets/changelog-github": "^0.6.0",
     "@changesets/cli": "^2.30.0",
+    "@earendil-works/pi-agent-core": "^0.79.0",
+    "@earendil-works/pi-ai": "^0.79.0",
+    "@earendil-works/pi-coding-agent": "^0.79.0",
     "esbuild": "^0.28.0",
     "typescript": "^5.7.0",
     "vitest": "^3.0.0"
   },
   "peerDependencies": {
-    "openclaw": ">=2026.5.22"
+    "openclaw": ">=2026.5.28"
   },
   "peerDependenciesMeta": {
     "openclaw": {
@@ -61,14 +63,14 @@
       "./dist/index.js"
     ],
     "compat": {
-      "pluginApi": ">=2026.5.22",
-      "minGatewayVersion": "2026.5.22",
+      "pluginApi": ">=2026.5.28",
+      "minGatewayVersion": "2026.5.28",
       "tested": [
-        "2026.5.22"
+        "2026.5.28"
       ]
     },
     "build": {
-      "openclawVersion": "2026.5.22"
+      "openclawVersion": "2026.5.28"
     }
   },
   "repository": {

package/skills/lossless-claw/SKILL.md

@@ -11,10 +11,11 @@ Start here:
 
 1. Confirm whether the user needs configuration help, diagnostics, recall-tool guidance, or session-lifecycle guidance.
 2. If they need a quick health check, tell them to run `/lossless` (`/lcm` is the shorter alias).
-3. If they suspect summary corruption or truncation, use `/lossless doctor`.
-4. If they want high-confidence junk/session cleanup guidance, use `/lossless doctor clean` before recommending any deletes.
-5. If they ask how `/new`, `/reset`, or `/lossless rotate` interacts with LCM, read the session-lifecycle reference before answering.
-6. Load the relevant reference file instead of improvising details from memory.
+3. If they are debugging lossless-claw behavior or failures, check the independent Lossless log before the shared OpenClaw gateway log.
+4. If they suspect summary corruption or truncation, use `/lossless doctor`.
+5. If they want high-confidence junk/session cleanup guidance, use `/lossless doctor clean` before recommending any deletes.
+6. If they ask how `/new`, `/reset`, or `/lossless rotate` interacts with LCM, read the session-lifecycle reference before answering.
+7. Load the relevant reference file instead of improvising details from memory.
 
 Reference map: