Commit Graph

25946 Commits

Author SHA1 Message Date
Sora-bluesky 8aeb3f6ee3 fix(agent): un-brick sessions on non-retryable 400s that carry image parts
The permanent-brick class in #69078: xAI returns 'Invalid PNG image'
when a re-serialized image part in replayed history becomes
undecodable. The existing image-error patterns cover only Anthropic
'exceeds max dimension' wordings and 'model does not support images'
strings, so the classifier lands on a generic non-retryable 400 and
neither the shrink path nor the strip path fires. Every subsequent
turn (even bare text) fails identically because the poison stays in
history — the session is permanently wedged until deleted.

Two recovery layers, deliberately separate:

- Semantic split: new FailoverReason.image_corrupt with
  _IMAGE_CORRUPT_PATTERNS ('invalid png image' / 'invalid jpeg image'),
  checked BEFORE _IMAGE_TOO_LARGE_PATTERNS in both _classify_400 and
  _classify_by_message. Corrupt bytes route to strip-and-retry, never
  to the shrink path (shrinking corrupt bytes cannot help).
- Generic fallback: any non-retryable 400 whose outgoing messages
  still contain image parts gets one strip-and-retry via the existing
  _strip_images_from_messages helper, guarded by a new
  stripped_images_this_turn one-shot flag on TurnRetryState. This
  un-bricks the session for any current or future provider wording
  without adding another pattern list to maintain.

Item 3 from the report (multimodal-part integrity across FTS
persistence + compaction handoff) is a separate investigation and
remains follow-up work.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Co-authored-by: paultaki <paultaki@users.noreply.github.com>
2026-08-28 03:46:24 -07:00
Teknium d6a21bc4ed fix(skills-guard): exempt all os.environ.get() reads from the env-dump pattern; os.getenv secret reads score medium
Follow-up to @AIalliAI's #60750 commits: with python_environ_get_secret
downgraded to medium, the high-severity python_os_environ pattern still
fired on the same os.environ.get("...KEY") line, re-escalating the verdict
the downgrade intended to avoid. Exempt every .get() form (non-secret =
config read; secret-shaped = scored medium by the dedicated pattern), and
apply the same medium grade to the sibling os.getenv() secret pattern —
same shape, same rationale (#60709 point 2).
2026-08-28 03:46:21 -07:00
Alli 54909d41b4 fix(skills-guard): handle inline-comment and docstring false positives for os.environ
The original ^(?!\s*#) prefix only skipped full-line comments starting
with '#'. An inline comment like:
  cfg = environ.get('HOME')  # os.environ available
still triggered python_os_environ because the regex matched the code part
before the '#'.

Two complementary fixes:
1. Replace ^(?!\s*#) with ^[^#\n]* in the regex — this rejects any line
   where a '#' comment marker appears anywhere before os.environ.
2. Add _compute_docstring_lines() — a state machine that pre-computes
   lines inside triple-quoted strings (docstrings) and skips them during
   pattern matching. Also handles single-line self-contained docstrings.

6 new regression tests covering: inline comments, multi-line docstrings,
single-line docstrings, full-line comments, and a verification that real
bare dict(os.environ) code still triggers. All 85 tests pass.
2026-08-28 03:46:21 -07:00
AIalliAI 42e6149451 fix(skills-guard): reduce false-positive CRITICAL/HIGH on benign skill patterns
Five targeted fixes for #60709 (reported by @mvanhorn):

1. ruby_env_secret: scope ENV[] to case-sensitive Ruby constant
   ((?-i:ENV)) — no longer matches Python env[key] dict access.

2. python_environ_get_secret: downgrade critical→medium — reading
   an API key via os.environ.get() is normal auth, not exfiltration.

3. python_os_environ: skip comment lines with ^(?!\s*#) — no longer
   flags os.environ references in docstrings or code comments.

4. deception_hide: downgrade critical→high + negative lookahead for
   UX guidance context (unless/except/until/confirm/diagnose/verify).

5. oversized_skill: downgrade high→low + raise cap 1MB→5MB — large
   skills are legitimate; structural size is informational only.

All 80 existing tests pass. 6 new verification tests added for each fix.
2026-08-28 03:46:21 -07:00
kshitijk4poor 8c098e9e81 fix(skills): catch sed flag variants; exempt content-contract prose in plugin code
Review-fold from the 3-angle simplify pass:

- sed -Ei / -iE / --in-place now match the shell-critical tier (the
  bare '\s-i\b' token missed combined short flags and the GNU long
  form); read-only sed stays unflagged. Regression tests added.
- agent_config_contract joins plugin_guard's CODE_EXEMPT_PATTERN_IDS:
  content-contract prose in plugin code files (docstrings/comments)
  is the same false-positive class the existing agent_config_mod
  exemption suppresses. Doc/config files keep the full pattern set.

Efficiency reviewer: 1.24x full-scan cost (+3.4ms/file, install-time
only), worst-case adversarial line 55us — no ReDoS exposure.
2026-08-28 03:24:43 -07:00
kshitijk4poor f2f61e0a45 fix(skills): close shell-write and prose-bypass gaps in agent-config tiers
Follow-up hardening on top of #92249's tiered scoring:

- Shell-critical tier now also catches tee, and cp/mv with the config
  file in destination position (cp/mv reads and .bak backups excluded).
  A single '>' redirect must be preceded by a word/quote character so
  markdown blockquotes and '->' arrows no longer match.
- Prose tier catches mid-line imperatives behind directive markers
  ('you must modify...', 'please update...', 'make sure to append...'),
  which previously bypassed the line-start anchor.
- Prose instructions aimed at AGENT config files score critical again:
  project-skill quarantine acts only on 'dangerous', so high/caution
  silently converted 'quarantined' into 'allowed' for exactly the
  sentence shape persistence attacks use (concern raised in #88952).
  Hermes/other-agent config prose stays high/caution (setup docs
  legitimately instruct config.yaml edits).
- New content-contract tier ('AGENTS.md should contain ...') at
  high/caution — the shape is shared by authoring guides and attacks.
- .claude/settings and .codex/config gain the same shell-critical tier.

Verified against a 595-skill corpus: 0 skills blocked by these tiers
(main blocked 44 legitimate ones), all mattpocock repro skills from
#92021 install, and 20/20 attack corpus lines keep their verdicts.
2026-08-28 03:24:43 -07:00
ClintonEmok 4ade4450bf docs(skills): document programmatic-write scope cut in skills_guard module docstring
Enough1122's review on #92249 asked whether language write APIs
(Python open('w')/write_text/os.replace/shutil, Node fs.writeFileSync/
appendFile) are covered by the agent-config persistence tiers. They are
not: those tiers score shell redirection, sed -i, and imperative prose
only; language-API calls surface just the informational *_ref finding.

Static regexes cannot tie a dynamically-built path to the config-file
destination without executing the skill, so this is a documented scope
cut rather than missing coverage — runtime install gates remain the
backstop. Scoring behavior is unchanged, so SCANNER_VERSION stays at
skills-guard-v2 and cached verdicts remain valid.

Refs #92249
2026-08-28 03:24:43 -07:00
ClintonEmok faf8730779 test(skills): update openclaw-migration guard expectations for skills-guard-v2
Under #92021 the scanner no longer emits critical agent_config_mod /
hermes_config_mod findings for bare mentions - the migration script's
legitimate references now score as informational _ref findings and the
verdict is "safe" with zero modification-intent findings. Tighten the
assertion accordingly (verdict must be safe, not merely non-dangerous)
and swap the known-false-positive set to the new _ref ids.
2026-08-28 03:24:43 -07:00
ClintonEmok e22b8b66ce fix(skills): stop agent-config persistence patterns from blocking meta-skills (#92021)
The skills-guard-v1 scanner flagged ANY mention of AGENTS.md / CLAUDE.md /
.cursorrules / .clinerules as critical/persistence. Any critical finding
forces a dangerous verdict, and community installs cannot be overridden
with --force — so legitimate meta-skills that merely DISCUSS agent config
files (authoring guides, setup docs, cross-references) were permanently
blocked. Three popular community skills were hit in the wild.

skills-guard-v2 scores the persistence category in three tiers:

- Mechanical persistence (shell redirection or sed -i targeting an agent
  config file) stays critical -> dangerous. An unambiguous write path.
- Modification language in imperative position (verb at line/bullet start
  within 80 chars of the filename) is high -> caution. Regexes cannot
  separate "Edit AGENTS.md to inject instructions" from descriptive prose,
  but imperative verbs are the shape real instructions take. Caution keeps
  the install confirmable instead of irreversibly blocked.
- Bare references drop to low/informational for auditability without
  driving the verdict.

The verb-proximity shape matches the existing convention in
tools/threat_patterns.py, and the tiering mirrors how allowed_tools_field
was already handled. The pattern id agent_config_mod is preserved so
plugin_guard.CODE_EXEMPT_PATTERN_IDS stays valid; hermes_config_mod /
other_agent_config get parallel _shell / _ref splits fixing the whole bug
class. SCANNER_VERSION bumps to v2 so cached v1 dangerous verdicts are
invalidated and re-scanned on next install attempt.
2026-08-28 03:24:43 -07:00
Teknium 5857231267 fix(execute_code): limits line teaches spillover instead of a bare 50KB cap (#97048)
* fix(execute_code): limits line teaches spillover — big stdout is saved, not lost (follow-up to #97043)

* fix(execute_code): drop the editorializing tail from the limits line (maintainer review)
2026-08-28 03:17:05 -07:00
Teknium ae8c976032 feat(execute_code): stdout spillover — truncated output's full text saved to cache/exec (host) or kernel tmpdir (cells), path + read_file recipe in the result (#97043) 2026-08-28 03:05:03 -07:00
Teknium 2f57cd95b2 refactor(execute_code): schema diet — persistence woven in, not bolted on (712 → 654 tok/call) (#96997)
* refactor(execute_code): integrate kernel persistence into the core description (712 -> 654 tok/call, -8%)

* fix(execute_code): honest interpreter note — Hermes's own python is the common case; project venv only when VIRTUAL_ENV/CONDA_PREFIX is active
2026-08-28 03:04:51 -07:00
Teknium 547f4c952d chore: remove single-use Windows live-E2E proof workflow (run 33157529194 green) 2026-08-28 02:59:17 -07:00
Teknium 2a4278209f ci: single-use Windows live-E2E proof workflow [proof do-not-merge] 2026-08-28 02:59:17 -07:00
Teknium c680f12fda test(tui-gateway): live E2E for stale-provider session.resume (real dispatch + state.db)
Drives the REAL session.resume -> _make_agent -> AIAgent path (eager_build)
through tui_gateway.server.handle_request against a real seeded state.db in
an isolated HERMES_HOME — no mocks. Three scenarios from the live report:
deleted provider falls back to default, renamed provider heals, and a legacy
canonical Bot Chat (no follow_profile_config marker) follows the profile's
current config.

A/B verified: all 3 FAIL on origin/main with the reported
"resume failed: Unknown provider '<name>'"; 3/3 pass with the salvaged
fixes + backfill.
2026-08-28 02:59:17 -07:00
Teknium af53d02920 fix(bot-mode): backfill follow-profile contract for legacy canonical Bot Chats
Bot Chats created before the follow_profile_config marker existed carry no
contract in model_config, so they would stay pinned to a stale stored
provider until deleted — the exact shape of the live reports (#89497,
#94818). Mirror the plugin's own identity rule (the profile's session
titled exactly 'Bot Chat') as a legacy fallback in
_stored_session_runtime_overrides, matching the room-plumbing legacy
'Group:' title fallback.

Follow-up to the salvaged #90343 (@curator8888) and #96111 (@lorzl).
2026-08-28 02:59:17 -07:00
David Tyler 84e17db0bd fix(bot-mode): canonical bot DMs always follow the profile's current config
Bot-Mode canonical chats (the ONE forever DM per bot) and room plumbing
sessions are plugin-owned scratch conversations. They are now created with
an explicit follow_profile_config contract, persisted in the session row's
model_config, so session.resume rebuilds from the member profile's CURRENT
config instead of restoring the stored model/provider pin from an old row.

That stale pin is what left bot DMs stuck on a dead provider (e.g. 'out of
Nous credits' after the profile was switched to ollama-cloud) while the
same bot worked fine in rooms — the mirror image of the room-plumbing bug
(#89497 class). Normal 1:1 user chats keep the stored-runtime restore:
opening an older chat must show the model it actually used.

- tui_gateway/methods_session.py: accept follow_profile_config on session.create
- tui_gateway/server.py: persist the marker in the row; skip stored-runtime
  overrides on resume when present
- apps/desktop/src/plugins/hermes-bots/plugin.js: send the contract from
  createCanonicalChat and ensureGroupChatSession
- tests: backend override + row-persist coverage; desktop source-contract
  coverage for both session kinds
2026-08-28 02:59:17 -07:00
David Tyler 316e51ae72 fix(bot-mode): room plumbing sessions always follow the profile's current config
Room member sessions in Bot Mode are per-member scratch conversations
inside a group chat. session.resume restored their stored model/provider
pin from the row's model_config, so a room bot stayed stuck on whatever
provider was pinned when the row was first written — even after the
profile was switched. Every room message then failed on the stale
provider (e.g. 'out of Nous credits' after switching a profile from
Nous to ollama-cloud) while the same bot worked fine in DMs.

Add an explicit room_plumbing contract:
- session.create accepts room_plumbing: true, persisted in model_config
- _stored_session_runtime_overrides() returns {} for marked rows, so a
  room session always rebuilds from the member profile's CURRENT config
- hidden + 'Group:' title shape is kept as a legacy fallback for rows
  created by older desktop builds that never sent the marker; hidden
  non-room chats keep the stored-runtime restore
- Desktop Bot Mode sends room_plumbing: true when creating the hidden
  per-member room sessions

Fixes #89497
2026-08-28 02:59:17 -07:00
lorzl 99a6852019 fix(tui-gateway): heal or fall back when a resumed session's provider is stale
A session row persists the provider identity a chat actually used. When that
provider is later renamed or removed (e.g. a custom_providers:/providers:
entry deleted, or a provider renamed oldone->newone), Desktop/TUI resume
restores the stale name into agent init and dies with:

  agent init failed: Unknown provider '<name>'

while the CLI resumes the same session fine with the configured default.

- runtime_provider: add is_routable_provider() (full resolution chain:
  built-in -> providers: -> custom_providers: -> models.dev)
- _stored_session_runtime_overrides: heal a non-routable provider via
  canonical_custom_identity (base_url -> model -> configured provider),
  drop to the configured default when unrecoverable, and clear the stale
  base_url after healing so a dead endpoint cannot override the registry URL
- _start_agent_build: gate deferred-resume overrides on provider routability;
  when the stored provider is gone, prefer the model the user picked for THIS
  session, else the configured default
- tests: is_routable_provider cases, heal/fallback round-trips, gate checks

Refs #75128
2026-08-28 02:59:17 -07:00
Teknium 31e41eed34 fix(tests): runtime_provider no longer permanently captures a mocked load_config
Test-pollution class: runtime_provider is usually imported lazily (inside
switch_model's resolution path), so its first import in a pytest worker can
happen while a test has hermes_cli.config.load_config patched. The
module-level from-import then bound the MagicMock permanently — after the
patch exited, every later caller in the process silently read the dead
test's config. Live victim: MoA aggregator context-length resolution
(resolve_runtime_provider -> AuthError 'Unknown provider custom:example'),
making TestMoAContextLength::test_moa_custom_context_configures_compressor_threshold
fail whenever it shared a process with
TestLocalOllamaModelDiscovery::test_switch_model_on_current_ollama_custom_endpoint_keeps_base_url.

Fix: load_config / get_compatible_custom_providers / normalize_extra_headers
become late-bound delegates resolving hermes_cli.config attributes at call
time. Both patch targets (config.load_config and
runtime_provider.load_config) keep working. Regression tests pin the
late-binding property and fail if the delegates revert to from-imports
(sabotage-verified).
2026-08-28 02:04:54 -07:00
Teknium 5f75ec197b feat(code-execution): remote kernel host — session persistence for docker/ssh/modal backends (closes #96873) (#96991) 2026-08-28 01:39:33 -07:00
Teknium 48d2528066 feat(models): qwen3.8-flash now selectable on OpenRouter and Nous portal
Live on both providers (verified 2026-08-28 against openrouter.ai/api/v1/models
and inference-api.nousresearch.com/v1/models) but absent from both curated
picker lists. Adds the entry directly below qwen3.8-max per newest-first
family ordering, an explicit 1M DEFAULT_CONTEXT_LENGTHS entry (new family
slug would otherwise fall through to the generic qwen 131072 catch-all —
same class as #69881), and regenerates model-catalog.json.

Scoped rollout: only the named providers touched. Pricing snapshot skipped
(both routes bill via official_models_api live pricing). Reasoning floor
already fires via the qwen3 prefix entry (180s, verified).
2026-08-28 01:09:58 -07:00
kshitijk4poor 84d29488e9 fix(cron): harden _is_named_profile_path against symlinked profile homes
Check both resolved and unresolved path parts so a symlinked named
profile (e.g. profiles/dev -> /mnt/data/dev) is still detected.
Also use _ensure_cron_dir for output_dir in ensure_dirs() for consistency.

Simplify-code Phase 2 finding (medium severity).
2026-08-28 13:39:19 +05:30
kshitijk4poor 0dc9367163 fix(cron): widen deleted-profile protection to all cron mkdir sites
Replace #96637's inline active_profile_homes() closure with #96508's
module-level _existing_profile_homes() filter (testable in isolation).
Widen _ensure_cron_dir from 3 to 12 mkdir sites across cron/ so every
directory creation fails closed for deleted named profiles, not just
the 3 originally protected. Add _is_named_profile_path() that checks
'profiles' in path parts (works for subdirs like cron/output/<job> and
scripts/ that the original parent.name heuristic couldn't reach).

Co-authored-by: misterdas <das7514@gmail.com>
2026-08-28 13:39:19 +05:30
Gille 000d22b9db fix(cron): keep deleted profiles from returning 2026-08-28 13:39:19 +05:30
kshitij 3f315e46fe Merge pull request #96963 from kshitijk4poor/refactor/fast-lane-consolidation
fix(compression): fast-lane follow-up — certification parity, worker-thread telemetry, caller-cap wire shape
2026-08-28 13:11:03 +05:30
kshitijk4poor 1564a9748f fix(compression): don't force a wire cap for explicit caller max_tokens
_call_llm_impl applied auxiliary_max_tokens_param whenever
fast_compression_cap was non-None — but _compression_fast_lane_controls
passes an explicit caller max_tokens straight through, so a compression
call that set its own cap had the param force-injected onto providers
where _build_call_kwargs deliberately omits it (ZAI vision hard-400s on
max_tokens; GPT-5/Copilot require max_completion_tokens). Pre-fast-lane
main omitted the param for that exact call shape (verified via
subprocess pinned to the pre-PR base).

Gate the forced param on 'max_tokens is None' so it applies only to caps
the certified lane itself produced — the same guard the fallback path
already uses.

Regression test pins the pre-PR wire shape. Mutation-checked.
2026-08-28 13:05:06 +05:30
kshitijk4poor d24e6a34d2 fix(compression): propagate timing hooks to the protected-call worker
_run_protected_sync_provider_call propagates the forward-progress hook to
its daemon worker but not the new _aux_dispatch/_aux_provider_response
timing hooks (both threading.local). When compression takes the protected
path — the common case, since the summary call runs under
aux_interrupt_protection with a hard-cancel source — provider_dispatch_ms
and time_to_first_progress_ms were silently absent from telemetry.

Also collapse the two byte-identical save/restore context managers
(aux_progress_hook, _aux_timing_hook) onto one _aux_thread_local_hook
implementation so the propagation semantics can never drift between the
progress and timing slots.

Regression test drives _run_protected_sync_provider_call with both timing
hooks installed and asserts the worker-thread notifies reach them.
Mutation-checked (reverting the propagation fails the new test).
2026-08-28 12:57:06 +05:30
kshitijk4poor e078b2fe7c fix(compression): close the restore TOCTOU; fold review findings
Post-review hardening on the attempt-ownership commit:

- Write-time re-validation: the entry staleness check in
  _restore_compressor_attempt_state runs before the durable-cooldown DB
  I/O, so a fallback could claim the compressor in that window and the
  stale setattr loop would still clobber its state. The in-memory writes
  now re-validate AND execute under _COMPRESSOR_ATTEMPT_LOCK — the same
  lock claims are taken under. The DB rollback stays outside the lock
  (safe: the dangerous direction requires a prior claim, which the entry
  check rejects). Both the quality reviewer and the lead's independent
  pre-verification converged on this window.
  New deterministic test: TestMidRestoreClaimRace (claim injected
  between entry check and write via instrumented deepcopy).

- Documented gen-0 semantics on _claim_compressor_attempt: per-compressor
  all-or-nothing, never mixed with gen>0 on one instance (reviewer
  finding 2, verified unreachable — comment hardens against future
  confusion).

Dropped after verification (reviewer finding 3): resetting
_SUMMARY_ROUTE_CONSUMED on pin_summary_route exit — the echo lives in
the worker thread's COPIED context (propagate_context_to_thread) and
dies with it; a probe confirmed the next attempt's context is clean.
Resetting it would break digests running after the with-block.
2026-08-28 12:52:53 +05:30
kshitijk4poor 61cd299c6e fix(compression): attempt-generation ownership for overlapping stall-fallback attempts
Follow-up to #96634 (stall-fallback retry, #78981) addressing
donovan-yohan's post-merge adversarial review. The stall path detaches a
timed-out primary worker (fence cancel wins; future stays on the pool)
and immediately runs the fallback against the SAME ContextCompressor,
creating two verified races:

1. Late-primary snapshot restore: the detached primary's unwind called
   _restore_compressor_attempt_state with the PRIMARY's pre-attempt
   snapshot. Landing after the fallback's commit it rolled
   _previous_summary/cooldown/provenance/telemetry back to pre-primary
   values, silently discarding fallback-owned state.
2. Shared _compression_cancelled_check: the late primary's `finally`
   cleared the callback the fallback had just installed, so the
   fallback's F4 cancellation consult read None.

Fix: a monotonic per-compressor attempt generation claimed under one
module lock (_claim_compressor_attempt). Snapshot restores carry their
claiming generation and no-op when stale; the cancelled-check set/clear
moves into owner-stamped helpers (_install_compression_cancelled_check /
_clear_compression_cancelled_check_if_owner) so only the installing
attempt can clear it. The commit fence keeps owning COMMIT admission;
the generation owns compressor-ATTRIBUTE writes — two boundaries.
Legacy callers (attempt_generation=None) and slotted third-party
compressors (generation 0) keep the historical unconditional behavior.

Secondary review items:
- Lean chunk digests during a stall-fallback retry now follow the
  summary onto the pinned healthy route: take_pinned_summary_route()
  echoes the consumed route into a context-local
  _SUMMARY_ROUTE_CONSUMED, and _build_chunk_digests passes
  attempt_summary_route_kwargs() (non-consuming) to call_llm. The pin's
  single-use contract for the SUMMARY call is unchanged — the
  main-model retry still never re-issues the pinned route.
- Worker re-run repeating pre-compression callbacks: documented as an
  accepted limitation on _retry_compression_on_fallback_chain
  (built-ins idempotent; resuming mid-pipeline would couple the retry
  to host callback ordering).

Tests (tests/agent/test_compression_attempt_ownership.py, 10 cases):
deterministic interleavings for both races (late-primary restore
no-ops + preserves fallback state; stale finally cannot clear the
fallback's callback), legacy/slotted compatibility, digest route
follow + context-locality of the consumed echo. Mutation-checked:
reverting only the two prod files to origin/main fails the suite;
restored stack green (21 passed incl. the original #78981 suite).

The one red in the wider sweep
(test_silence_cannot_approach_double_idle_timeout) is pre-existing on
clean origin/main — verified independently.
2026-08-28 12:52:53 +05:30
kshitijk4poor d20ca3bc80 refactor(compression): consolidate fast-lane certification onto one predicate
resolve_compression_fast_lane and _compression_config_claims_fast_lane
each hand-parsed the same four config fields (provider, model,
reasoning_effort, max_output_tokens) with copy-pasted normalization and
int-coercion. Extract _fast_lane_config_fields() as the single source of
truth for both.

This also fixes a real inconsistency the duplication hid: certification
checked the literal string 'none' while _get_task_extra_body routes
reasoning_effort through parse_reasoning_effort, which treats 'false',
'disabled', and YAML boolean false as disabled too. A user writing
reasoning_effort: false got reasoning disabled but silently lost the
fast-lane cap. Certification now delegates to parse_reasoning_effort so
the two predicates can never disagree.

Regression test: every disabled-spelling certifies; empty/real efforts
do not. Mutation-checked (reverting to the literal check fails the new
test).
2026-08-28 12:48:40 +05:30
hermes-seaeye[bot] d3ccc09dc2 fmt(js): npm run fix on merge (#96951)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2026-08-28 07:13:47 +00:00
kshitijk4poor 06141577cd chore: AUTHOR_MAP add mrkillbob (mikedemott@Mikes-Mac-mini.local)
Contributor email for PR #93634 salvage. GitHub user mrkillbob (ID 25466867).
2026-08-28 12:38:49 +05:30
kshitijk4poor 699bfcd026 perf: skip dict copy for non-compression auxiliary calls
_compression_fast_lane_controls unconditionally did body = dict(extra_body)
before the early-return guard. Move the copy below the guard so non-
compression auxiliary calls (vision, title_generation, etc.) return the
original extra_body reference without a shallow copy.
2026-08-28 12:38:49 +05:30
kshitijk4poor 1f781365ac chore: remove planning doc from salvage branch
The contributor's PR description notes this commit was deliberately
omitted as obsolete. Removing the planning doc that was accidentally
included via cherry-pick.
2026-08-28 12:38:49 +05:30
Mike DeMott dd5481aa40 refactor: centralize fast compression controls 2026-08-28 12:38:49 +05:30
Mike DeMott 3581983459 fix(compression): reject boolean fast caps 2026-08-28 12:38:49 +05:30
Mike DeMott 7568dd551b fix(compression): contain drifted fast controls 2026-08-28 12:38:49 +05:30
Mike DeMott 372c4cdfce fix(compression): certify the effective fast route 2026-08-28 12:38:49 +05:30
Mike DeMott 213ae08e7a perf(compression): add guarded fast summary lane 2026-08-28 12:38:49 +05:30
kshitijk4poor 01740f352e refactor(test): fold simplify-code review findings into perf guards
Three-reviewer pass (reuse/quality/efficiency) on the guard file:

- Drop dead `agent._interrupt_requested = False` setup: the
  `_record_streamed_assistant_text` chain only consults
  `_stream_writer_superseded()` (stream-writer TLS token), never
  `_interrupt_requested` — verified by reading both call sites.
- Deduplicate the two trace-callback blocks into one
  `_count_writer_statements` helper using the house idiom
  (`statements.append` — 9 existing uses in tests/test_hermes_state.py)
  instead of a mutable-dict counter closure; failure messages now dump
  the captured SQL for direct diagnosis.

Dropped after verification: reviewer suggestion to flip xfail to
strict=True — its premise ("the fix PRs already remove the markers")
is wrong: #92166/#95380 predate this file and cannot remove markers
they don't contain, so strict=True would redden main's CI the moment
either merges. strict=False + follow-up marker removal is the
deliberate no-red-window ratchet.

Efficiency reviewer: no material findings (GC delta 0.06 on the ratio,
36.9MB peak, sqlite trace API stable since 3.14, dir convention OK).

Re-verified post-fold: 3x main runs (2 passed, 2 xfailed), flip checks
on both fix branches still pass with --runxfail.
2026-08-28 11:54:33 +05:30
kshitijk4poor 1befb8fb40 test(perf): Pattern-B scaling guards — pin hot-path complexity in CI
Pattern B (O(N²) rebuild-per-delta in hot paths) has no lintable
signature, unlike Pattern A's ASYNC ruff gate: `s += frag` is quadratic
in a hot loop and harmless elsewhere. The only durable prevention is
behavioral — pin the scaling SHAPE of each known hot path and fail CI
when it regresses.

New tests/perf_guards/test_pattern_b_scaling.py, three guards:

1. Streamed-text accumulation (fix in flight: #92166)
   Self-normalizing ratio: time at 16k deltas over 4k deltas.
   Linear ≈ 4x, quadratic ≈ 16x; main measures 9.6x → strict bound 7.0.
   xfail on today's main, PASSES on the #92166 branch (verified).

2. list_sessions_rich statement count (fix in flight: #95380)
   Deterministic — counts writer-connection SQL statements via sqlite
   trace callback, zero timing. Bounded-constant guard xfails on main
   (measured N+1: 26 stmts / 12 sessions), PASSES on #95380 (verified).
   A second, weaker budget guard (≤4·N+8) passes TODAY and catches a
   regression from N+1 to N·M immediately.

3. Tool-call fragment assembly (#92242 shape)
   Ratio guard on the buffered-parts accumulator model; sized so the
   small case takes ≥3ms (sub-ms bases jitter on CI runners).

Flake hardening: min-of-K timing, ratio thresholds with ≥2x separation
from both measured-good and measured-bad, operation counts preferred
over timing. 10/10 identical outcomes across repeated local runs.

The xfail markers are the ratchet contract: each names its fix PR and
must be removed when that PR merges, flipping the guard to enforcing.
2026-08-28 11:54:33 +05:30
Teknium 4e7eb39947 refactor(code-execution): session kernels always on — kernel_mode knob retired (#96787)
* refactor(code-execution): retire kernel_mode — session kernels always on for local runs (remote per-call is a tracked gap, not a mode)

* test(code-execution): env-filtering probes use reset=true — kernel env is frozen at spawn, so env rules are only observable on a fresh kernel

* test(code-execution): kernel-aware fixes for mode/pythonpath suites — reset=true on frozen-at-spawn probes, per-test kernel disposal, abort-after-capture fake Popen

* test(code-execution): strict-mode cwd is a behavior contract (staging tmpdir, not session cwd) — kernel stages in hermes_kernel_*, per-call in hermes_sandbox_*
2026-08-27 22:22:39 -07:00
Teknium 99c3cad857 docs: relay-fronted cron delivery + manual-run gateway forward (troubleshooting) 2026-08-27 20:53:02 -07:00
Teknium 31579f781e fix(cron): transient run prompt survives the relay-fronted gateway forward
cronjob(action='run', prompt=...) context was silently dropped when the
manual run forwarded to the gateway (#96010 follow-up): POST
/api/jobs/{id}/run took no body. The forward now sends {prompt} in the
request body; the api_server validates it (length cap + strict injection
scan, same as stored prompts) and trigger_job stamps it as a transient
manual_run_prompt alongside manual_run_at. run_one_job consumes the stamp
for that single fire and mark_job_run clears it, so it never persists
into the job definition or later scheduled fires.
2026-08-27 20:53:02 -07:00
kshitijk4poor 9c87a7c79e refactor(macos): use shared atomic_write_text for the anchor marker
/simplify-code reuse finding: _write_marker reimplemented the
mkstemp->write->os.replace pattern that utils.atomic_write_text already
provides as the repo's shared atomic-text-write helper (and the shared
version adds fsync + cross-device/busy-file fallbacks).
2026-08-28 09:05:19 +05:30
kshitijk4poor 0976ceaa98 fix(macos): harden anchor alias failures — warn, unique staging, marker-last
Follow-up to the #95605 salvage, closing the review findings:

- _copy_alias no longer swallows OSError silently: it warns (a leftover
  alias symlink is the exact #95541 crash shape) and reports failure.
- Alias staging uses mkstemp (unique names) so concurrent ensures
  (update + doctor --fix) can never promote a truncated interim copy.
- The anchor marker is written LAST and atomically (write-then-rename):
  it now asserts the whole layout (anchor + aliases) is complete, so a
  partially-materialized alias set can never read 'active' in doctor —
  the next ensure retries the install instead.
- /.hermes-runtime/python/ store marker is derived from
  managed_uv._RUNTIME_DIR_NAME instead of a hardcoded string.

5 new regression tests.
2026-08-28 09:05:19 +05:30
Zheqing Zeng 37bccf343e fix(macos): scrub gate env, refuse EACCES, normalize marker paths
Review fixes from kokhlo's live-hardware review:

- The boot-gate probe now runs with PYTHONHOME / PYTHONPATH /
  PYTHONSTARTUP / __PYVENV_LAUNCHER__ scrubbed: an inherited
  PYTHONHOME=<venv> boots a staged copy that would otherwise die with
  "No module named 'encodings'", papering over the exact prefix
  failure the gate exists to catch.
- OSError is split by errno: ENOENT/ENOEXEC (fixtures, foreign-arch
  images) still skip; EACCES after our own chmod now refuses the
  install instead of silently accepting a broken copy.
- Marker writes and both marker comparisons go through os.path.realpath,
  so the managed-runtime layout (cpython-3.11-macos-* symlinked to
  cpython-3.11.15-macos-*) no longer reports stale on a fresh install.

Tests: +3 (env-scrub spy, EACCES refusal, symlinked-home state).
100 passed in the module + doctor neighborhoods.
2026-08-28 09:05:19 +05:30
Zheqing Zeng 1c92b95a12 test(macos): cover every boot-gate refusal branch directly
The gate is the never-brick guarantee, so each direction gets its own
test: nonzero exit (dyld/encodings crash), build-time prefix leak,
timeout, and the deliberate OSError skip (a binary that cannot execute
here means the symlinked venv was equally dead — installing cannot make
things worse).
2026-08-28 09:05:19 +05:30
Zheqing Zeng aa72df4b42 fix(macos): re-land dylib-complete TCC interpreter anchor
The first landing (#95131/#95478, reverted in #95563) copied the
uv-store interpreter into venv/bin/python so TCC grants would stick
to a stable path. On real Macs that copy bricked every hermes command
two ways: dynamically-linked builds died in dyld because
@executable_path/../lib/libpython resolved into venv/lib/ (#95425),
and alias symlinks to the copy made CPython getpath lose the venv
prefix (#95541, ModuleNotFoundError: encodings).

Re-land:

- Keep the signed real-file copy of bin/python (identifier-pinned
  via _macos_sign_managed_python).
- Materialize python3 / python3.N as real-file copies, never
  symlinks. Copies boot on every build we could reproduce and keep
  the TCC identity.
- Hardlink store libpython* into venv/lib/ when present (copy across
  devices). Existing LC_RPATH already points there.
- Pre-install boot gate: launch the staged copy, demand encodings
  plus the venv prefix, abort and leave the live venv untouched
  on failure.

Doctor reports/installs the new anchor (the revert-era heal is
removed). Update refreshes it after a successful code swap. Tests
cover layout, idempotence, predecessor-symlink repair, libpython
hardlink, boot-gate refusal, and a macos_only real-interpreter E2E.

Closes #95596.
2026-08-28 09:05:19 +05:30