ce67ffb70e
feat(agents): add tool-free isolated completion ( #114343 )
...
* feat: add isolated pure-inference completion
* fix(google): block ambient system prompt writes
* docs: refresh generated map
2026-07-30 13:24:45 -05:00
0090549188
fix(agents): refresh Claude live system prompts ( #116292 )
...
Co-authored-by: William Faris Chesnutt <farischesnutt@gmail.com >
2026-07-30 23:40:40 +08:00
9eae43bd37
refactor(infra): move exec approvals into the shared SQLite state DB ( #114063 )
...
* 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
2026-07-26 06:39:23 -04:00
82d1a03f25
refactor(agents): move implicit-main fallback into load-time roster injection ( #112678 )
...
* refactor(agents): require explicit roster defaults
* feat(onboard): create named first roster agent
* refactor(agents): remove runtime main fallbacks
* style(agents): apply roster refactor formatting
* refactor(agents): finish roster-only runtime sweep
* fix(doctor): migrate legacy main session sqlite
* fix(doctor): harden roster session migrations
* fix(onboard): commit first agent atomically
* fix(config): support empty-roster analysis
* fix(agents): preserve legacy main state during creation
* fix(setup): materialize baseline agent roster
* fix(agents): harden legacy default transfer recovery
* fix(agents): simplify roster-only legacy compatibility
* fix(agents): preserve staged first-agent entries
* fix(config): migrate persisted implicit-main rosters
* fix(config): preserve staged empty rosters
* fix(agents): finalize roster-only upgrade paths
* fix(sessions): close legacy main migration outcomes
* fix(config): migrate legacy roster markers at load
* fix(sessions): preserve roster upgrade history
* refactor(sessions): restore lean legacy main compatibility
* fix(setup): prepare first-agent credentials before publish
* fix(config): stabilize roster snapshot migration
* refactor(sessions): shrink legacy main compatibility
* fix(agents): restore roster compatibility fidelity
* fix(sessions): preserve divergent legacy history
* refactor(agents): narrow roster-only scope
* fix(config): isolate roster migration
* test(agents): align roster-only fixtures
* fix(agents): keep main agent undeletable
* fix(agents): harden roster migration invariants
* fix(agents): close setup and audit scope gaps
* fix(cron): scope session reaper throttles by agent
* fix(agents): preserve scoped owner precedence
* fix(config): preserve authored config ownership
* fix(setup): keep default workspace and roster in sync
* fix(setup): preserve default entry workspace on bare runs
* fix(agents): adapt roster rebase to keyed entries
* fix(agents): honor both roster representations
* fix(agents): route roster reads through shared helpers
* fix(config): preserve canonical roster writes
* fix(cron): resolve dynamic default for session reaper
* fix(agents): close dynamic default migration gaps
* fix(agents): align scoped session ownership
* fix(sessions): preserve legacy main directory casing
* fix(agents): align cron and legacy auth ownership
* fix(setup): provision the committed default workspace
* fix(cron): align scoped ownership and reaping
* fix(cron): treat blank agent ids as absent
* fix(cron): retain configured session-store owners
* fix(agents): repair roster-aware CI boundaries
* fix(cron): preserve scoped ownership resolution
* fix(agents): preserve rosterless maintenance paths
* fix(agents): propagate roster ownership through runtime boundaries
* fix(agents): preserve roster ownership across runtime paths
* fix(agents): harden roster diagnostics and legacy routing
* fix(agents): remove redundant diagnostic import
* test(agents): type CLI policy fixture explicitly
* fix(config): preserve canonical roster mutation identity
* fix(doctor): read canonical agent rosters consistently
* fix(config): resolve compound roster unsets safely
* fix(config): finalize main-session reconciliation
* fix(doctor): read canonical session state safely
* fix(sessions): preserve current visibility alias
* fix(config): track roster include provenance
* test(config): type roster provenance cases
* fix(config): refine roster include ownership
* fix(agents): preserve staged roster invariants
* test(config): align fixtures with explicit roster ownership
* test(node-host): preserve optional plan typing
* fix(config): preserve authored roster projections
* test(config): keep raw roster fixtures explicit
* test(config): normalize rosters at runtime fixtures
* fix(config): protect authored roster ownership
* fix(agents): require explicit session ownership
* fix(agents): enforce scoped roster ownership
* fix(sessions): merge fixed-store agent partitions
* fix(agents): harden roster ownership boundaries
* fix(config): reject ambiguous roster projections
* fix(sessions): preserve persisted store ownership
* fix(sessions): keep collision diagnostics additive
* fix(security): scan malformed roster workspaces
* test(config): align snapshot fixtures after rebase
* test(agents): use explicit roster fixtures
* fix(config): harden roster diagnostic boundaries
* fix(sessions): isolate fixed-store agent databases
* test(agents): type malformed default markers
* refactor(sessions): extract store collision resolution
* test(system-agent): split oversized setup coverage
* style(system-agent): format split setup suite
* fix(sessions): preserve promoted store ownership
* fix(sessions): derive scoped owner before target
* fix(sessions): preserve explicit sqlite ownership
* fix(agents): restore roster compatibility across CI
* fix(agents): enforce roster-owned runtime boundaries
* fix(agents): satisfy default lookup lint
* test(sessions): split known-owner coverage
* fix(state): satisfy path identity lint
* fix(agents): preserve malformed roster safety boundaries
* fix(agents): restore roster compatibility at runtime boundaries
* fix(config): satisfy roster boundary type checks
* fix(agents): preserve roster ownership across runtime probes
Setup inference probes now execute as the configured roster owner. Malformed agent-prefixed session rows are intentionally omitted by the fail-closed visibility contract rather than normalized by tests.
* fix(agents): satisfy session list owner lint
* fix(agents): preserve roster-owned runtime boundaries
Restore shared logical rows for exact SQLite session locators while keeping their physical database owner separate. The ownership regression test now constructs an explicit sole-owner database directly instead of relying on first-touch capture, matching the intentional shared-store contract.
* fix(sessions): preserve multiply owned exact stores
* fix(sessions): restore runtime owner boundaries
Keep incognito sentinels agent-owned, fold default-agent approvals into the global snapshot, and preserve the configless legacy-main CLI policy fallback. Also repair the existing CLI watchdog test lifecycle so the compact shard observes its timeout without an unawaited assertion or async timer stall; product behavior is unchanged by that test-only fix.
* test(ci): align owner-scoped fixtures
These assertions are unchanged. The fixtures now declare the intended non-default runner, expose the session-key constant imported by production status code, and select the main approvals bucket explicitly on Windows.
* fix(agents): close final roster ownership gaps
2026-07-24 22:38:09 -07:00
a325467557
fix(cron): preserve scheduled caps across CLI and worker upgrades
2026-07-24 01:24:03 -05:00
8287e21445
fix(cron): enforce caps across CLI backends
2026-07-24 01:24:03 -05:00
3023d69fd8
feat(gateway): carry reviewer-only detail on plugin approvals ( #113027 )
...
* feat(gateway): carry reviewer-only detail on plugin approvals
* chore(protocol): regenerate projections for plugin approval detail
* chore(plugin-sdk): budget reviewer-detail surface additions
2026-07-23 06:28:23 -07:00
a229456f48
feat(agents): relay Claude native tool requests as Gateway approvals ( #112918 )
...
* feat(agents): relay Claude native tool requests as Gateway approvals
* fix(agents): keep approval relay types local and refresh docs map
2026-07-23 03:21:08 -04:00
f6818ae5ea
refactor(media): hydrate and prune images from structured facts, delete marker-text parsing ( #112913 )
...
* refactor(media): hydrate and prune images from structured facts, delete marker-text parsing
Embedded, CLI, host-context replay, and plugin-harness image hydration now
consume the persisted/runtime MediaFact carrier instead of parsing
OpenClaw-authored [media attached:]/[Image: source:] marker text. The
prompt-text attachment parser family (splitPromptAndAttachmentRefs, leading/
trailing marker extractors, ref-count consumption, text-based prune scrubbing)
is deleted; explicit user-authored paths and file URLs still hydrate.
- Described-image suppression is a declared MediaFact.hydrationSuppressed
field carried across copy/reprojection boundaries; the runtime symbol
plumbing is removed.
- Persisted facts plus mediaImageLayout are authoritative at every hydration
entry point; image blocks reconcile by structural fact index, never
byte/MIME equality.
- Offloaded attachments (image and non-image) persist their media://inbound
claim-check alias so ownership-aware history pruning can scrub markers
after persistence; factless pre-MediaFact rows keep narrow legacy redaction.
- History pruning removes facts, layout, and block provenance structurally
before replay.
Prompt bytes are golden-equal throughout; goldens from the fact-carrier PR
are untouched. 1,303 tests across 44 files; delegated check:changed green
(run 29980534186, pinned to the declared merge-base).
* fix(media): fail missing inline image slots
* test(agents): carry offloaded CLI retry image as a structured fact
The retry-image reliability test modeled an offloaded attachment with only
its [media attached:] marker text; offloaded attachments are carried as
structured media facts and marker text is presentation-only, never parsed
for hydration.
2026-07-23 03:11:16 -04:00
c4bd3f5b2e
refactor: consolidate CLI runner test fixtures ( #112779 )
...
* test: consolidate cli runner fixtures
* test: satisfy CLI fixture guards
2026-07-22 15:28:48 -07:00
1a42e005fb
fix(anthropic): forward selected profiles to Claude CLI ( #112458 )
...
* fix(anthropic): forward Claude CLI auth profiles
* fix(system-agent): inject CLI auth route stores
* fix(claude-cli): pass profile credentials by descriptor
* fix(anthropic): repair selected profile CI coverage
* fix(anthropic): preserve profile owner validation
* test(system-agent): preserve selected profile fixtures
* test(system-agent): narrow selected profile fixture
* test(system-agent): resolve profile store merge
* fix(anthropic): forward profiles to node Claude runs
* fix(system-agent): reconcile profile route projection
* test(system-agent): thread profile store through projection
* fix(anthropic): make selected profile authoritative
* fix(system-agent): type auth setup failures
* fix(system-agent): type setup auth failures
* style: format Claude profile maintenance
* fix(anthropic): keep gateway credentials off nodes
* fix(anthropic): clear ambient auth for selected profiles
* fix(anthropic): secure paired-node Claude auth
* fix(node-host): type Claude fd spawn streams
* style(node-host): satisfy Claude spawn lint
* fix(process): capture exit before secret delivery
* fix(anthropic): preserve node-native Claude auth
2026-07-21 23:27:37 -06:00
edecdbd05e
refactor(config): config-surface reduction tranche 3 — product consolidations (review request) ( #111527 )
...
* refactor(config): consolidate media model lists
* refactor(config): unify memory configuration
* refactor(config): consolidate TTS ownership
* refactor(config): move typing policy to agents
* refactor(config): retire product-level config surfaces
* refactor(config): share scoped tool policy type
* chore(config): refresh generated baselines
* fix(config): honor agent typing overrides
* fix(config): migrate sibling config consumers
* refactor(infra): keep base64url decoder private
* fix(config): strip invalid legacy TTS values
* chore(config): refresh rebased baseline hash
* fix(doctor): route legacy messages.tts.realtime voice to talk during tts move
* refactor(config): polish final layout names
* refactor(config): freeze retired tuning defaults
* feat(config): add fast mode default symmetry
* refactor(config): key agent entries by id
* docs(config): update final layout reference
* test(config): cover final layout migrations
* chore(config): refresh final layout baselines
* fix(config): align final layout runtime readers
* fix(config): align remaining readers
* fix(config): stabilize final layout migrations
* fix(config): finalize config projection proof
* fix(config): address final layout review
* docs(release): preserve historical config names
* fix(config): complete keyed agent migration
* fix(config): close final migration gaps
* fix(config): finish full-branch review
* fix(config): complete runtime secret detection
* fix(config): close final review findings
* fix(config): finish canonical docs and heartbeat migration
* fix(config): integrate latest main after rebase
* refactor(env): isolate test-only controls
* refactor(env): isolate build and development controls
* refactor(env): collapse process identity indirection
* refactor(env): remove duplicate config and temp aliases
* docs(env): define the operator-facing allowlist
* ci(env): ratchet production variable count
* fix(env): remove stale provider helper import
* fix(env): make ratchet sorting explicit
* test(env): keep test seam in dead-code audit
* test(env): cover ratchet growth and boundary; document surface budgets
* docs(config): document tier-eval consolidations
* docs(config): clarify speech preference ownership
* test(memory): align retired tuning fixtures
* refactor(memory): freeze engine heuristics
* refactor(config): apply tier-eval tranche
* refactor(tts): move persona shaping to providers
* refactor(compaction): move prompt policy to providers
* test(config): align hookified prompt fixtures
* chore(deadcode): classify test-only exports
* chore(github): remove unused spawn helper
* chore(deadcode): classify queue diagnostics
* chore(deadcode): remove unused lane snapshot export
* chore(plugin-sdk): ratchet consolidated surface
* fix(config): integrate latest main after rebase
2026-07-21 20:28:43 -07:00
783a5d21cf
refactor(config): purge numeric tuning knobs behind built-in defaults ( #111382 )
2026-07-19 07:35:45 -07:00
60ed910639
fix(diagnostics): own run activity listener lifecycle ( #110312 )
2026-07-18 03:29:06 +01:00
ddb4ed9a47
feat: trace Claude Code CLI model calls ( #108304 )
...
* feat: trace Claude CLI model calls
* fix: align Claude CLI trace hierarchy
* fix: distinguish model request and turn telemetry
* fix: satisfy diagnostics CI checks
* fix: distinguish agent-turn telemetry
---------
Authored-by: Alex Knight <15041791+amknight@users.noreply.github.com >
2026-07-18 06:23:49 +10:00
9497450511
refactor: eliminate dead-export baseline ( #108376 )
...
Burn the grandfathered unused-export baseline to zero and enforce a hard-zero Knip gate.
2026-07-15 17:05:07 +01:00
d49e12eb36
refactor(agents): privatize CLI runner and live auth helpers ( #108282 )
...
* refactor(agents): internalize live auth helpers
* refactor(agents): privatize CLI runner test seams
* chore(deadcode): shrink CLI runner export baseline
2026-07-15 04:50:35 -07:00
81d634be50
test: speed CLI capture suites ( #107993 )
2026-07-14 21:16:34 -07:00
a6a0716486
feat(setup): rename Crestodian to OpenClaw system agent
...
User-facing name is now OpenClaw (the system speaks); internal code name is
system-agent. Gateway methods crestodian.* -> openclaw.chat/openclaw.setup.*,
agent tool -> openclaw, reserved agent ids openclaw + retired crestodian.
openclaw setup routes: onboarding flags -> onboard, -m/--yes -> system agent,
bare configured interactive -> OpenClaw chat, unconfigured -> onboarding.
Hidden crestodian CLI and /crestodian TUI aliases kept; docs moved to
docs/cli/openclaw.md with redirect stub. macOS/Android strings in lockstep.
Refs #107237
2026-07-14 11:03:02 -07:00
f81f9d8570
chore: enforce max-lines suppression ratchet ( #107315 )
...
* ci: enforce max-lines suppression ratchet
* chore: prune stale max-lines suppression
* fix: close max-lines ratchet enforcement gaps
* fix: harden max-lines ratchet checks
* fix(ci): satisfy max-lines ratchet checks
* style: format max-lines declarations
* fix(ci): match oxlint suppression grammar
* test: isolate max-lines git fixtures
* chore: prune resolved max-lines debt
* test: skip newline path fixture on Windows
* fix: harden max-lines suppression ratchet
* chore: refresh max-lines baseline
* fix: close max-lines ratchet bypasses
* fix: derive ratchet base from PR merge tree
* fix: support older Git in staged ratchet
* fix: align max-lines declarations and baseline
* chore: refresh max-lines baseline for current main
* fix: exclude generated wizard locales from max-lines
* chore: prune resolved max-lines debt
2026-07-14 09:27:02 -07:00
6d31c162f9
fix: keep root Claude live sessions usable ( #107606 )
...
* fix(agents): support Claude live sessions as root
* chore: leave release notes to release workflow
2026-07-14 08:46:58 -07:00
f167089d61
feat(nodes): continue Claude catalog sessions on paired nodes via streaming CLI agent runs ( #105833 )
...
* feat(gateway): stream node.invoke progress with idle timeouts and cancel
* feat(node-host): opt-in approval-gated claude cli agent run command
* feat(agents): run node-placed claude-cli turns through streaming node invoke
* feat(anthropic): continue claude catalog sessions on advertising nodes
* docs(nodes): document paired-node claude agent runs and continuation
* chore(plugin-sdk): admit spawn-invocation helper into surface budget
* fix(nodes): harden streaming invoke and node run policy from review findings
* test(node-host): prove multi-token tool-policy argv fails closed
* fix(nodes): allowlist cancel event for apps and break node-host type cycle
* chore(protocol): regenerate Swift gateway models for node invoke progress
* chore(node-host): drop test-only export surface flagged by deadcode ratchet
* style(node-host): merge invoke-types imports
* refactor(nodes): split node agent run modules to satisfy the LOC ratchet
* chore(protocol): regenerate android gateway models and sync export baseline
* chore(plugin-sdk): admit spawn-invocation helper and inherited deprecated drift into surface budget
* chore(plugin-sdk): tolerate inherited agent-core deprecated export
* fix(anthropic): drop node param parser orphaned by upstream catalog split
* ci: sync unused-export baseline with healed main
* test(agents): adopt renamed cli backend prepare helper
* style(node-host): format claude run spawn invocation
* ci: retrigger checks
* chore(plugin-sdk): pin surface budgets to healed-base counts
* fix(protocol): restore untyped lazy validators after upstream style change
2026-07-13 07:39:22 -07:00
ed57c6cd88
fix(agents): honor claude-cli contextTokens setting ( #93198 )
...
* fix: honor claude-cli configured context tokens
* test: cover claude-cli native compaction budget
* test: cover ordinary claude-cli budget persistence
* test: migrate claude-cli budget fixtures to session accessor
* fix(agents): carry CLI provider into context budget
* fix(agents): guard optional CLI context metadata
* fix(agents): align Claude CLI native compaction budget
Co-authored-by: 杨浩宇0668001029 <yang.haoyu@xydigit.com >
* fix(agents): apply selected CLI context cap
Co-authored-by: 杨浩宇0668001029 <yang.haoyu@xydigit.com >
* fix(agents): preserve prepared CLI context budgets
Co-authored-by: 杨浩宇0668001029 <yang.haoyu@xydigit.com >
* fix(agents): resolve prepared CLI context ownership
Co-authored-by: 杨浩宇0668001029 <yang.haoyu@xydigit.com >
* test(agents): type CLI context budget cases
Co-authored-by: 杨浩宇0668001029 <yang.haoyu@xydigit.com >
---------
Co-authored-by: Peter Steinberger <steipete@gmail.com >
2026-07-13 01:58:44 -07:00
fb48df4a0f
fix(claude-cli): max-turn stops omit run and recovery context ( #94130 )
...
* fix(anthropic): handle max_turns stop reason and surface retry-limit details to users
* fix(anthropic): classify max_turns as length-limited
* fix(anthropic): preserve max_turns diagnostics
* fix(anthropic): preserve max_turns side-effect warning
* fix(claude-cli): max-turn stops omit run and recovery context
* fix(claude-cli): preserve max-turn terminal boundaries
* refactor(claude-cli): centralize terminal error conversion
Co-authored-by: zhang-guiping <zhang.guiping@xydigit.com >
* fix(claude-cli): warn before retrying max-turn failures
Co-authored-by: zhang-guiping <zhang.guiping@xydigit.com >
* fix(claude-cli): preserve max-turn stops through wrappers
Co-authored-by: zhang-guiping <zhang.guiping@xydigit.com >
* fix(claude-cli): prioritize terminal max-turn output
---------
Co-authored-by: 张贵萍 <zhangguiping@zhangguipingdeMac-mini.local >
Co-authored-by: Peter Steinberger <steipete@gmail.com >
2026-07-12 19:57:34 -07:00
1b313dc4d4
chore(tooling): enforce indexed access in core tests ( #105375 )
...
* test(tooling): enforce indexed access in core tests
* test(tui): constrain theme environment overrides
* test(doctor): enforce indexed access in migration fixtures
2026-07-12 14:29:52 +01:00
96f0983a85
fix(onboarding): skip setup for configured gateways and require inference first ( #102883 )
...
* fix(crestodian): keep onboarding RPCs restart-safe
* fix(profiles): isolate approval state migrations
* fix(crestodian): bypass configured gateway setup
* test(crestodian): type onboarding mocks
* fix(onboarding): require inference before Crestodian
* fix(onboarding): enforce verified inference handoff
* fix(macos): reset setup on gateway endpoint edits
* chore(i18n): refresh native source inventory
* fix(gateway): keep socket on request cancellation
* test(packaging): require workspace templates
* fix(onboarding): bind setup to verified inference
* fix(onboarding): align inference gate contracts
* fix(crestodian): classify concurrent policy rejection
* test(crestodian): expect registry restoration
* fix(onboarding): bind setup to configured gateways
* fix(codex): preserve startup phase deadlines
* test(crestodian): match fail-closed policy ordering
* test(onboarding): assert bound gateway handoff
* fix(codex): bind runtime resolution to spawn cwd
* test(crestodian): assert policy rejection order
* fix(cli): preserve gateway routing across restarts
* fix(macos): fail closed during gateway edits
* test(macos): cover gateway route generation races
* chore: keep release notes out of onboarding PR
* fix(ci): refresh onboarding generated checks
* style(swift): align gateway channel formatting
* fix(ci): refresh plugin SDK surface budgets
* fix(ci): resync native string inventory
* refactor(swift): split gateway channel support
* test(doctor): isolate plugin compatibility registry
* test(macos): isolate gateway onboarding fixtures
* test(macos): assert gateway lease health ordering
* fix(codex): reconcile computer-use startup changes
2026-07-11 10:25:14 -07:00
5c260940bd
fix: prevent CLI MCP children from widening sandbox tools ( #103822 )
...
* fix(gateway): bind CLI MCP grants to context
* chore: leave release changelog ownership intact
2026-07-10 17:44:29 +01:00
59e95fe3fd
feat: support GPT-5.6 Ultra across OpenClaw and Codex runtimes ( #98021 )
...
* feat: support GPT-5.6 Ultra across agent runtimes
Co-authored-by: J Cai <anyech@gmail.com >
* fix: keep harness projections discovery-free
* fix(codex): mirror V2 native subagent tasks
* chore: refresh plugin SDK surface budgets
* test: expose Ultra wire effort proof
* test(cron): avoid hoisted mock initialization race
---------
Co-authored-by: Peter Steinberger <steipete@gmail.com >
2026-07-10 15:23:24 +01:00
28ce79a94a
perf(test): resolve mocked live-run cancellation
2026-07-10 07:52:44 -04:00
b752384586
fix(agents): recover claude-cli warm-stdin continuity when no native transcript is written (reseed candidate not dropped) ( #96841 )
...
* fix(agents): recover claude-cli warm-stdin continuity when no native transcript is written
The headless warm-stdin claude-cli backend (liveSession: "claude-stdio")
never writes a native transcript, so the post-turn flush probe always
fails and the missing-transcript reuse path drops the bound session id.
Part 1 (cli-runner.ts): scope the non-destructive binding behavior to
warm-stdin sessions so they keep their binding instead of clearing it
every turn.
Part 2 (attempt-execution.ts): on a missing transcript, clear the stored
binding (no stale --resume) but still return the bound id as the reuse
candidate so prepare can re-detect the missing transcript and arm
raw-transcript reseed. Returning undefined starved reseed and lost
warm-stdin continuity.
Adds/updates regression coverage in attempt-execution.cli.test.ts and a
complementary reseed test in prepare.test.ts.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com >
* test(agents): align warm-session continuity coverage
* test(agents): keep cli live-session mock complete
* fix(agents): respect stateless CLI session mode
* style(agents): keep session candidate guard focused
* test(agents): preserve minimal CLI runner fixtures
* fix(agents): preserve exact Claude warm sessions
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com >
Co-authored-by: Peter Steinberger <steipete@gmail.com >
2026-07-10 03:48:18 +01:00
4af2ca0a4d
fix(agents): extend live CLI no-output watchdog to the blocked-tool floor
2026-07-09 14:26:05 +05:30
969efdac4f
fix(agents): make run liveness evidence-based
2026-07-09 14:26:05 +05:30
f53346944d
feat: correlate native search outcomes in audit history ( #98704 )
...
* feat: correlate native search outcomes in audit history
Metadata-only audit ledger for agent runs and tool actions in the shared
state DB: stable event identity, closed action/status/error vocabularies,
one-way-hashed tool-call ids, never-inferred terminal outcomes for native
web-search (explicit completed/failed only; otherwise unknown), bounded
retention, audit.list gateway RPC and openclaw audit CLI. Squashed from
the 82-commit audit stack for replay onto current main.
* feat(audit): add audit.enabled config gate (default on)
The metadata-only audit ledger records by default: an audit trail enabled
only after an incident cannot explain the incident, and the rows are
strictly less sensitive than the transcripts every install already
stores. audit.enabled=false stops new writes at the gateway subscription
seam; audit.list and openclaw audit keep serving existing records until
they expire. Documented in the configuration reference, protocol page,
and CLI reference.
* fix: repair full-matrix CI findings after rebase
- break the dynamic-tools/dynamic-tool-execution import cycle by
extracting resolveCodexToolAbortTerminalReason into a leaf module
- restore main's session-worktree protocol exports lost in the
index.ts auto-merge
- register the audit event writer worker as a knip entry point
- docs table formatting; subagent wait-cancellation test scoped to its
audit intent (outcome + timing) and advanced past main's new
lifecycle-timeout retry grace
2026-07-06 12:30:12 +01:00
3c2316c43c
fix(agents): preserve streamed assistant text when Claude CLI result event is empty ( #90450 )
...
* fix(agents): preserve streamed Claude CLI replies
Co-authored-by: Luiz Antonio Busnello <246086438+totobusnello@users.noreply.github.com >
* fix: preserve Claude live commentary boundaries
* docs(changelog): stabilize Claude CLI entry
* chore(changelog): defer Claude CLI entry
---------
Co-authored-by: Peter Steinberger <steipete@gmail.com >
Co-authored-by: Luiz Antonio Busnello <246086438+totobusnello@users.noreply.github.com >
2026-07-06 02:18:20 +01:00
b332918a22
test(agents): cover soft cli session drift reuse
2026-07-03 22:16:43 -07:00
11441aefbe
fix(agents): classify quiet Claude live exits
...
Fixes #98946
2026-07-02 09:31:56 -07:00
d18817019f
fix(agents): unify cli stream output classification
...
Fixes #98896
2026-07-01 23:14:58 -07:00
d02d8b1ea5
fix(agents): keep cli live retry artifacts
...
Fixes #98897
2026-07-01 22:39:24 -07:00
52cabf877d
fix(agents): recover cli context overflow sessions
...
Fixes #98897
2026-07-01 22:39:24 -07:00
8939c10d77
fix(agents): handle claude live write timeout rejection
...
Fixes #98894
2026-07-01 22:39:18 -07:00
a5a8d992ae
fix(claude-cli): return updatedInput in can_use_tool allow response ( #98665 )
...
Claude Code >= 2.1.156 tightened its PermissionResult schema so an
approved can_use_tool control_response allow branch must carry an
updatedInput record. The Claude live-session bridge answered with the
deprecated { behavior: "allow" } shape and no updatedInput, so the CLI
rejected every approval-gated native tool call (Bash, WebFetch, and
AskUserQuestion among them) with a ZodError before the tool ran.
Echo the tool input back unchanged as updatedInput on the allow branch,
matching the shape the Claude Code 2.1 binary expects. The deny branch
and all other behavior are unchanged.
Fixes #95171 .
2026-07-01 08:11:38 -07:00
92b283da84
refactor(process): remove unused orphan reconciliation API
2026-06-23 01:13:13 +08:00
a2675756b8
test(agents): stabilize Claude live spawn coverage
2026-06-22 09:16:31 +08:00
92d1f04de3
refactor(agents): drop duplicate internal aliases
2026-06-19 18:33:56 +08:00
7bdea54a82
fix: scope cli runner env helpers
2026-06-18 15:14:09 +01:00
519eb3d088
refactor(agents): remove stale Claude CLI compat wrapper
2026-06-18 14:21:26 +08:00
b2f7d9ebc8
fix: allow Gemini CLI file edits
2026-06-17 03:31:14 +01:00
c6d7d85763
fix: parse Gemini CLI stream output
2026-06-17 03:31:14 +01:00
81c9bd3997
fix: add CLI empty response diagnostics
2026-06-17 03:31:14 +01:00
0bc384fc03
fix: keep CLI env diagnostics opt-in
2026-06-17 03:31:14 +01:00
Josh Avant