* feat(agents): rank Tool Search with BM25 over names, descriptions, and parameters
Ranking was case-insensitive substring matching with hand-tuned weights, which
failed in ways that made tools unreachable rather than merely mis-ordered:
- "scheduling" found nothing against a tool described "Schedule a recurring
task" — no stemming.
- "read" ranked spreadsheet_open, because "sp-read-sheet" contains it.
- A non-English query tokenized to zero terms, and the scorer returned 1 for
every entry, so the model received an arbitrary alphabetical slice of the
catalog presented as a ranked result.
Replace it with Okapi BM25 over a tokenizer that splits on Unicode word
boundaries, drops stopwords, and collapses light English inflection. Index
parameter names and descriptions too, which Codex (bm25 crate) and the Claude
API tool-search tools both do; a query like "repository" now reaches a tool
whose description never says it. Underscore-joined names index as both the whole
name and its parts.
A small query-expansion table bridges intent to description vocabulary ("look up
the price" -> search/web), which pure lexical overlap cannot do. It holds only
generic capability words, never plugin or vendor names.
Empty queries now score nothing instead of everything, and both `tool_search`
and the code-mode bridge tell the model to query in English, so the degenerate
case is steered away from rather than silently mishandled.
Untrusted schemas are still never traversed: parameters are indexed only for
`openclaw` entries, matching the boundary compactToolSearchCatalogEntry already
enforces by reporting MCP and client inputs as "unknown".
* fix(agents): undouble inflected consonants and state the real English contract
Autoreview caught two defects in the new stemmer and its documentation.
"running" stripped to "runn", which can never meet "run", so a tool named
task_runner described "Running tasks" became unreachable for the query "run" —
a regression the substring scorer did not have. English doubles the final
consonant before -ing/-ed/-er, so undo that, while keeping doubles that belong
to the root (call, process, off, buzz).
The English-only claim was also stronger than the code: the tokenizer keeps
Unicode letters, so a non-English query yields terms and can match. Rejecting
them would make a catalog that legitimately names a tool in another script
permanently unreachable, so the behavior stays and the wording now matches it —
catalogs are written in English, so other languages usually match nothing, and
the model is asked to query in English for that reason rather than because the
input is filtered. The previous test only proved an unrelated document scored
zero, so it is replaced by one that asserts each half directly.
* fix(agents): split camelCase, discount expansions, and stop trigger collisions
Three defects in the new tokenizer, all found by autoreview and all reproduced
before fixing.
`splitWords` lowercased before looking for boundaries, so `readFile` produced
only `readfil` and the natural query `read file` could not meet it. MCP catalogs
commonly use camelCase, and the old substring scorer matched those. Split case
transitions before lowercasing.
Expansion terms carried the same BM25 weight as words the caller typed, so
`weather` — which expands to search/web — could rank a general web tool above
the exact weather tool by matching two terms instead of one. Expansions are a
guess about how the catalog words a capability, so they now score at 0.35, and a
term the caller actually wrote keeps full weight even when an expansion repeats
it.
Trigger matching ran the document stemmer, which collapsed unrelated vocabulary:
`news` became `new`, so "open a new issue" silently acquired a web-search intent.
Triggers now normalize by singularization only, which leaves `news` intact, and
the table lists `reminder` explicitly rather than relying on the stemmer to
reach it.
* fix(agents): normalize -ies plurals and tier literal matches above expansions
`repositories` stemmed to `repositori` while `repository` stayed put, so the two
never met; `-ies` now normalizes back to `-y` in both the document stemmer and
the expansion triggers, which also lets `directories` and `memories` reach their
intended groups instead of stalling at `directorie`.
The 0.35 expansion discount is not sufficient on its own to keep a literal match
ranked first: BM25 sums per term, so a common literal like `weather` carries
little IDF while a short document collecting two rare expansions can outscore
it, and a small result limit then drops every tool matching the typed word.
Literal overlap is now reported per hit and ranked as a tier ahead of score, so
the discount orders within a tier rather than trying to carry the invariant.
Verified by running the ranker over the failing inputs rather than reasoning
about them; the first attempt at the trigger fix silently did not apply because
the formatter had reflowed the function, which the matrix run caught.
* fix(agents): keep non-plural -s words out of the plural stem rule
"news" stemmed to "new", which literal-matched every "Create a new ..." tool;
because literal overlap now outranks expansion-only hits, a search for news
returned creation tools instead of the web tool the query meant.
The collision is not unique to news — "status", "canvas", and "alias" are all
ordinary tool vocabulary here and all lose their meaning under the same rule, so
the exemption covers that class rather than the single reported word.
* fix(agents): restore the exact-name tier and keep `get` searchable
Two signals the old substring scorer had and BM25 alone does not.
It gave an exact name/id match +20, which flattening everything into one
document removed: querying a known tool name could rank a shorter entry that
merely mentions the word above the tool itself, and the result limit would then
drop the tool asked for. Exact name/id is now a sort tier ahead of literal
overlap.
`get` was in the stopword list, but it names real operations in a tool catalog.
Discarding it made `search("get")` empty and reduced "get issue" to "issue",
where a shorter delete_issue or update_issue entry can win on length
normalization. Capability verbs stay indexed.
* fix(agents): keep acronyms whole, both -ies readings, and stopword-named tools
* refactor(agents): keep the ranking types and entry-text helper module-local
Knip's hard-zero unused-export gate flagged WeightedTerm, RankedDocument,
LexicalIndex, and toolSearchEntryText: nothing outside their own modules
imported them. Narrowing the surface is what the repo asks for anyway.
The untrusted-schema test now drives ToolSearchRuntime.search instead of
calling toolSearchEntryText directly, which proves the boundary through the
real entry point. Verified it still bites: removing the source gate makes it
fail with "client properties must remain deferred".
* feat(ui): durable session board face and dashboards index
Board face lived only in client-side boardSessionViews, capped at 50 entries,
so the preference never followed the user to another device, evicted as
sessions accumulated, and could not be seen as a set.
Persist it as SessionEntry.boardFace, which rides the existing entry_json blob
and so needs no SQLite schema change or version bump. Expose it on the session
list row and add it to the sessions.patch write-scope allowlist alongside label,
pinned, and archived: setting your own view preference is user-level chat
organization, not policy. Unknown patch fields still fail closed to
operator.admin.
Generic navigation now reads the stored face, so the sidebar and session list
open a thread on the face you left it on. boardSessionViews keeps only
activeTabId and reopenDockByTab, which are genuinely per-device.
Add /dashboards listing threads whose preferred face is dashboard. Filtering
runs server-side in filterSessionEntries before pagination, because the client
holds only a capped page and a client-side filter would silently omit
dashboards.
* test(protocol): assert the pre-rename face param is rejected
The gateway-protocol validator test still passed the pre-rename 'face' key,
which the closed schema rejects. Use boardFace, and pin the old name as a
negative case so it cannot silently return.
* chore(protocol): regenerate Swift bindings and docs map for boardFace
Adding boardFace to the sessions schema changes two committed generated
artifacts: the Swift gateway models (pnpm protocol:gen:swift) and the docs map
(pnpm docs:map:gen), which now lists the dashboards index section.
* feat(runtime): allow Bun runtimes that provide node:sqlite
* fix(process): drop execa buffer encoding under Bun spawn (Bun rejects non-spawn options)
* chore(process): cite oven-sh/bun#36049 in bun spawn workaround
* docs(install): bun with node:sqlite can run openclaw; bun install workspace caveat
* fix(process): clear execa buffer encoding under Bun without mutating read-only options
* refactor(prompt): plain inbound context labels with a provenance marker
Replaces trust-worded inbound context labels ("(untrusted metadata)",
"(untrusted, for context)") with plain labels plus a fixed provenance
marker suffix appended to every OpenClaw-injected context header.
Detection keys on the marker, not label text, so strippers stay correct
across UI, TUI, replay, /trace segmentation, memory recall, and the Swift
chat preprocessor. Drops sanitizeInboundSystemTags in favor of the marker
boundary plus trusted system-prompt narration.
Renames the untrusted-named plugin SDK context identifiers to
channel-provenance names, keeping deprecated aliases registered for
removal after 2026-09-08.
Adds `openclaw doctor --fix` migrations that rewrite legacy inbound
labels in stored SQLite transcripts and purge legacy envelope-
contaminated LanceDB recall rows.
* fix(ci): resolve gate failures for plain inbound context labels
- doctor sqlite readers: open read-only connections via openNodeSqliteDatabase
so the Kysely connection-boundary guardrail holds; unexport the now-internal
transcript snapshot type (Knip unused-export gate).
- compat registry: split the record table into registry-records.ts and
plugin-sdk-subpath-records.ts. The new compat record pushed registry.ts past
the 700-line oxlint cap; suppressions are disallowed, so follow the existing
sibling record-module pattern. Public exports and PluginCompatCode literals
unchanged.
- acp-runtime test: assert current finalization behavior (newline normalization
only). The bracket de-fang and System: rewrite it expected were removed with
sanitizeInboundSystemTags; forged system lines are neutralized at the
system-event queue, the single chokepoint feeding the System:-per-line render.
- regenerate docs_map and the plugin SDK API baseline manifest.
* fix(prompt): harden inbound context label migration and drop in-band sanitizer
Review follow-ups on the plain-label + provenance-marker change:
- Remove src/security/system-tags.ts. Rewriting inbound text to neutralize
look-alike `System:`/`[System]` markers corrupted legitimate user text and is
not a real injection boundary; role separation plus external-content wrapping
is. Explicit product decision, recorded at the system-event queue.
- Narrow the LanceDB legacy-row purge so it cannot delete benign memories. It
now requires a complete known legacy sentinel line, a legacy label followed by
a fenced JSON body, or the complete legacy external-content header. The prior
predicates matched ordinary prose such as `Notes (untrusted metadata):`, and
deletion is irreversible.
- Make explicit-empty canonical ChannelStructuredContext win over the deprecated
alias via a present/absent result instead of collapsing `[]` to undefined.
- Keep `\r?` in the active-memory doctor rule. It is the only rule spanning the
header's line break, migrated assistant rows skip newline normalization, and
without it the marked-header replace wins and the body strips to empty. Added
a CRLF regression test.
- Fix stale comments that described removed behavior, and cover the Swift
prose-block strip path.
Claude-Session: https://claude.ai/code/session_01WNzsPddQmxy9Y7jKD4wAxH
* feat(ui): path-based session and dashboard URLs
* docs(ui): document control UI URLs
* fix(ui): finalize session path routing
* feat(ui): anchor session URLs on stable keys
* docs(ui): clarify stable session URL identity
* fix(ui): resolve session prefixes with full prefix and pagination
Query sessions.list with the full supplied prefix instead of an eight-char
truncation, and paginate up to a bounded page count so longer disambiguation
links resolve instead of being reported ambiguous. Zero strict-prefix matches
now fall through to literal-key resolution rather than rendering an empty
ambiguity view.
Also document the ~dot/~dotdot segment escape: peer ids reach session keys
trimmed and lowercased only (src/routing/session-key.ts), so a literal '.' or
'..' segment is reachable and browsers would normalize it away.
* fix(ui): synchronize committed session routes
* test(ui): split native shell host coverage
* fix(ui): thread configured mainKey through session URL builders
Reserved-set disambiguation needs the operator-configured mainKey at runtime, so
thread it from agentsList through every session path builder and the ClickClack
control URL. Unambiguous non-hex single-segment rests now resolve literally while
short-id-shaped rests still fail closed, which restores ClickClack channel
compatibility detection and control-link reconciliation.
Also bound prefix-resolution retries, preserve catalog thread identity, and keep
draft state on ambiguous candidate links.
* fix(ui): repair session URL CI integration
* perf(ui): lazy-load session route resolution
* perf(ui): isolate session prefix resolution
* perf(ui): defer session path parsing
* perf(ui): defer session navigation startup
* fix(ui): preserve first-run and literal session navigation
* fix(ui): satisfy session routing type and export gates
* fix(clickclack): preserve unscoped control link agent
* style(ui): satisfy chat page line limit
* refactor(ui): move chat page helpers to owning modules
* fix(ui): preserve destination session route identity
* fix(ui): preserve agent identity in session routes
* fix(ui): escape dots in literal session path segments
encodeURIComponent leaves periods intact, so a literal key segment like
channel:release.js produced /chat/main/channel/release.js. In-app navigation is
intercepted by the SPA, but a refresh, an external link, or a ClickClack link
would be served as a static asset request and never reach the app.
pathForWorkboardBoard already escapes dots for this reason; mirror it in both the
session URL contract and the ClickClack encoder, route the agent id through the
same segment encoder, and pin the case in both shared vector tables.
* fix(ui): stop bootstrap after teardown race
* refactor(ui): centralize session navigation targets
* fix(ui): consume bootstrap teardown abort
* fix(ui): canonicalize configured main session routes
* fix(ui): preserve distinct session references
* fix(ui): redirect released session query links
* fix(ui): make bootstrap teardown abort-safe
* refactor(infra): move exec approvals into the shared SQLite state DB
Delete the file-runtime exec-approvals store (exec-approvals.json + .lock
sidecar machinery) on both runtimes and make the reserved
exec_approvals_config singleton row canonical. Doctor owns the one-time
import with claim/verify/receipt discipline; runtime fails closed with a
doctor instruction while un-migrated legacy state exists. The wire CAS
contract, socket semantics, and gateway auth-token derivations are
unchanged. Kills the #113929 lock-contention bug class structurally and
nets around -2.9k lines.
* fix(infra): green CI gates and retire file-era exec approvals tests
Break the migration-type import cycle with a leaf contract, regenerate the
plugin-SDK API and native i18n baselines for the intentional surface change,
drop unused exports, and replace the macOS file-era approvals test suite with
SQLite-backed behavior coverage per the obsolete-internals test policy.
* chore: green max-lines ratchet, native i18n baseline, and unused-export scan
* fix(plugins): import adopted catalog history into the OpenClaw transcript
Import native Pi and OpenCode history when a catalog session is adopted so the OpenClaw transcript reflects the full conversation immediately.
Mark imported user rows with mirrorOrigin because they are transcript mirrors, not new external turns. This preserves watcher deduplication and prevents adopted history from being reported as fresh upstream input.
* refactor(plugins): split catalog history importer
Keep the session catalog contract module leaf-like by moving transcript mutation into a dedicated runtime module. This breaks the registry-to-transcript import cycle while preserving the Plugin SDK entrypoint and importer behavior.
* build(deps): remove npm shrinkwrap; mirror pnpm lock into transient package locks
npm 12 removed shrinkwrap (command + tarball/root loading). Delete all 82
committed npm-shrinkwrap.json files and stop publishing lockfiles; keep
pnpm-lock.yaml as the single reviewed dependency boundary. The generator
becomes scripts/generate-npm-package-lock.mjs and feeds plugin bundling via
a transient package-lock.json + npm ci (works on npm 11 and 12). Tarball
validation treats the published 2026.7.2 beta train as a shrinkwrap
transition; self-update npm detection now uses install topology instead of
the shipped shrinkwrap.
* fix(deps): repair lint, deadcode, and test-type lanes for the npm 12 migration
- sort integrity comparisons with an explicit comparator (oxlint)
- keep resolveBunGlobalNodeModules module-local (knip unused-export gate)
- model npm pack --json as npm<=11 array / npm 12 name-keyed object
- default calver destructuring in the tarball test fixture
* feat(ui): gateway-owned sidebar section order with hover-only drag grips
* fix(ui): repair sidebar section order CI gates
* fix(protocol): keep sectionOrder optional for older gateways
Native Swift clients ship separately from gateways, so new clients must decode older gateway responses that omit sectionOrder.
* fix(ui): reconcile gateway section ordering with main
* fix(state): allow lazy tables in v5 maintenance
* refactor(ui): retire prefs session section order in favor of gateway-owned order
Supersedes #113948 by deleting the unreleased ui.prefs.sessionSectionOrder key and its browser plumbing. Gateway SQLite sidebar_sections remains the single canonical store.
* chore(compat): date the annotated deprecation families and expose removal-pending debt
Adds 10 dated compat-registry records for shipped deprecated surfaces that
previously had @deprecated annotations only (removeAfter 2026-10-01 per the
approved 60-day window), extends the beta5 session-store record with the
package-root aliases, surfaces removal-pending records with due dates,
blockers, and readers in the plugin boundary report, and converts undated
cleanup obligations (CI timing job, zizmor excessive-permissions, staged
Codex CI auth, memory-host legacy state dir, agent-DB schema-ladder floors)
into dated markers visible to the weekly dated-TODO sweep. Metadata,
diagnostics, and docs only - no runtime behavior change.
* fix(compat): keep extension src test paths out of core string literals
The core import guardrail forbids extensions/*/src/ path shapes in core
production string literals; cite the package-root google-meet test and the
registry test for official-plugin-export-aliases instead.
* fix(providers): stop reading operator's real config in openai live tests
* fix(qianfan): refresh onboarding default off deprecated deepseek-v3.2
* fix(providers): satisfy live-test config type
The policy/data-handling-redaction-disabled check could never emit a finding:
scanPolicyDataHandling records the sensitiveLoggingRedaction evidence with a
hardcoded value true, and the finding builder only fired on value !== true.
Redaction is unconditional in src/logging/redact.ts, which hardcodes tools mode
and reads only redactPatterns, so no config can turn it off.
Delete the check, its finding builder, check id, and the validateOnly fix class
that existed solely for it. Keep the public
dataHandling.sensitiveLogging.requireRedaction policy key: it is a policy.jsonc
contract, it still drives openclaw policy compare baseline strictness, and its
shape stays validated.
Make the key's satisfied status explicit instead of silent: the policy rule
declares satisfiedByInvariant pointing at the evidence source policy state
records (oc://openclaw.invariant/logging/redaction), which openclaw policy check
emits in dataHandling evidence and the attestation. A metadata test asserts every
rule names either its checks or its invariant, never both and never neither, and
that policy state actually emits the declared source.
Also drop the stale logging.redactSensitive entry from the policy config coverage
manifest; that config key is retired.
* feat(ui): drag custom sidebar groups between built-in session zones
* test(ui): remove stale split ratio fixture
* perf(ui): raise startup budget baseline for session-section ordering pref
* feat(plugins): support additional upstream session kinds
* feat(acpx): monitor upstream Pi sessions
* feat(opencode): monitor upstream OpenCode sessions
* fix(opencode): ignore hidden user text activity
* refactor(plugins): share upstream echo filtering
* fix(opencode): preserve marker tuple narrowing
* refactor(opencode): detect upstream turns via event_sequence cursor
Replace timestamp/grace-window completeness heuristics with OpenCode's own
per-session event_sequence.seq cursor, which advances inside the same
transaction as the message/part projections. Detects change instead of
inferring completion, so staged part writes can no longer drop a turn.
Handles cursor regression from OpenCode migrations that clear event_sequence,
suppresses compaction replay and summary re-publish, and restricts human-turn
classification to visible non-synthetic text.
* docs(plugins): record why the adoption coordinator requires complete
* fix(opencode): align cursor proof with strict types
* fix(acpx): keep Pi JSON parser private
* fix(control-ui): drop stale Create PR row after merge and surface PR chips mid-turn
* test(gateway): split session-PR branch tests to satisfy max-lines
* test(gateway): drop unused GitContext export from test support
The update_plan checklist tool has been default-on with no model gating for a
while, so keeping it under tools.experimental misrepresented it and kept a
container key alive for a single boolean. Rename it to a plain boolean
tools.updatePlan (camel-cased tool id, matching tools.agentToAgent and
tools.toolSearch), keeping the !== false default-on semantics.
The strict schema rejects the retired key, so openclaw doctor --fix now lifts
tools.experimental.planTool to tools.updatePlan and deletes the emptied
container. Registration also drops five parameters that were never read
(agentSessionKey, agentId, modelProvider, modelId, pluginToolAllowlist).
Docs: config-tools.md dropped a stale claim that the tool defaults off behind a
strict-agentic GPT-5 rule that no longer exists, experimental-features.md no
longer lists the tool and now cites the real Codex app-server floor (0.143.0,
per MIN_CODEX_APP_SERVER_VERSION).
Config baseline core count drops 2307 -> 2306.
Sensitive log redaction became unconditional and `logging.redactSensitive`
was retired from the config schema, but the policy doctor still classified
`dataHandling.sensitiveLogging.requireRedaction` as an automatic repair that
would have written the retired key back into config.
Remove the automatic repair (check-id registration, patch branch, and the
`enableSensitiveLoggingRedaction` writer), drop the `logging.redactSensitive`
config target, and reclassify the check as `validateOnly`. The check keeps
evaluating the policy declaration against the redaction invariant; it just no
longer claims a fixable config target. Refresh the policy docs, the finding
fix hint, and a stale comment in `src/logging/redact.ts`.
* docs: correct retired cron/audit config keys, cron failure-alert default, memory recall default, and tool-search telemetry claims
- configuration-reference: cron block documented cron.webhook and cron.failureDestination, both retired by the config-surface reduction tranches (58452de711, edecdbd05e); the cron schema is strict so a copied snippet is rejected. Document only the live keys and note the doctor --fix migrations.
- configuration-reference: root-level audit block is retired; canonical path is logging.audit (src/config/zod-schema.root-shape.ts).
- configuration-reference: cron.failureAlert.after default is 2, not 3 (src/cron/service/failure-alerts.ts).
- memory-config: rememberAcrossConversations defaults on for personal installs (packages/memory-host-sdk/src/host/config-utils.ts), matching the canonical table earlier in the page.
- tool-search: telemetry records catalogSize, per-source counts, and search/describe/call counts, and only on tool_search_code results. No byte accounting exists in the runtime.
* docs: retire remaining references to removed cron, audit, and logging config keys
Sweep follow-up to the previous commit, covering the same bug class in the pages that still contradicted it.
- cron-jobs/cli-cron: global cron.failureDestination is retired; the destination fields now live on cron.failureAlert (src/config/zod-schema.root-shape.ts, merged by legacy-config-migrations.runtime.retired.ts:379). Per-job delivery.failureDestination bullets left intact.
- gateway/audit, cli/audit, gateway/protocol: root-level audit.* is retired; canonical path is logging.audit.*.
- logging: logging.redactSensitive is retired (dead-config-keys.test.ts:198; removed by legacy-config-migrations.runtime.tier-eval.ts:12). resolveConfigRedaction hardcodes DEFAULT_REDACT_MODE = tools, so redaction is unconditional. Also documented that redactPatterns replaces the defaults on the log path (redact.ts:419) while tool payloads always merge them.
- logging: consoleStyle accepts only pretty|json (zod-schema.root-shape.ts:106); compact remains the automatic non-TTY rendering style (logging/console.ts:40) but is no longer settable, and doctor maps a stored one to pretty.
- security: security --fix no longer touches redaction and the logging.redact_off audit check is retired (src/security/audit-loopback-logging.test.ts asserts it never fires).
* chore(docs): regenerate docs map after retired-key cleanup
