CLI: the VAD monitor's playback cut and the record-key interrupt both
mark the latch; the chat path prepends the note via the existing
_prepend_note_to_message channel. TUI: _tts_stream_stop() grows a
user_barge flag (False for /voice off — a mode change isn't an
interruption; no-op when speech already finished), the VAD monitor
marks on cut, prompt.submit accepts interrupted:true from clients, and
_run_prompt_submit pops the latch into the run message.
mark_speech_interrupted() / take_speech_interrupted(): a one-shot,
TTL'd (120s) latch plus SPEECH_INTERRUPTED_NOTE. Barge-in paths mark
it when they cut live speech; the next turn's submit path pops it and
prepends the note to the model-bound message — API-call local, never
persisted, so history and prompt caching are untouched.
/api/audio/speak-stream WebSocket: one socket + one Web Audio clock per
reply. The renderer feeds raw LLM deltas as they arrive; the server
cuts sentences with the shared chunker and streams int16 PCM back while
generation continues — speech overlaps generation with no per-sentence
connection or synthesis gaps. Falls back to the POST data-URL path for
old backends / non-chunked providers.
Barge-in runs a MediaRecorder on the monitor's stream the whole time
playback is live (rotated while quiet to bound pre-roll); talking over
the agent cuts playback and the complete utterance goes straight to
transcription and submit. /api/audio/transcribe returns 200/"" for
no-speech results so quiet turns re-listen instead of toasting a 400.
Turn deltas feed a per-turn TTS pipeline; the post-complete speak_text
call survives only as a fallback. session.interrupt, /voice toggles,
and new turns cut in-flight speech. VAD barge-in emits
voice.interrupted at detection, then the captured interruption goes out
as voice.transcript — the same event the TUI already submits as a
spoken turn.
The streaming gate broadens from ElevenLabs-only to any provider that
passes check_tts_requirements(). In continuous voice mode a mic monitor
runs during playback: talking over the agent cuts TTS at detection
while the monitor keeps recording, then the captured interruption is
transcribed and queued as the next turn (process_loop's auto-restart
stands down while the capture owns the mic). New voice.barge_in config
key, default true.
listen_for_speech(): sustained-RMS speech detection with the noise
floor calibrated against audible playback (speaker bleed doesn't
self-trigger). With capture=True it keeps a rolling pre-roll buffer and
records through to a silence endpoint, so the interruption is
transcribed from its first syllable — detection alone loses the opening
words to the detector's sustain window plus the mic re-open.
transcribe_recording() maps no_speech provider results to a successful
empty transcript (silence, not an error).
An STT provider that hears no words is reporting silence, not failing.
ElevenLabs/xAI empty-transcript errors now carry no_speech so live
voice loops can re-listen quietly instead of surfacing an error on
every pause.
StreamingTTSProvider ABC + registry + resolve_streaming_provider() —
ElevenLabs (pcm_24000) and OpenAI (response_format=pcm) stream chunked
PCM; every other provider keeps its configured voice and falls back to
per-sentence sync synthesis. SentenceChunker is the one incremental
cutter every surface shares: sentence-boundary cuts on the delta
stream, <think> blocks stripped even when split across deltas, short
fragments merged forward so they never stall as tiny clips.
Removes Homebrew and PyPI wheel/sdist as Hermes distribution paths while
preserving the supported source, Docker, and Nix workflows.
Changes:
- Removes the Homebrew formula, PyPI publish workflow, sdist manifest
(MANIFEST.in), and wheel/sdist release-attachment logic from scripts/release.py.
- Keeps setuptools metadata and entry points required by editable installs
and Docker/Nix builds, but adds a setup.py guard that rejects wheel/sdist
builds outside a sealed Nix derivation (HERMES_NIX_BUILD=1).
- Removes pip/Homebrew install detection, PyPI update checks, the pip
self-update path, the deprecation-banner state, the postinstall subcommand,
wheel data-directory fallbacks in agent/i18n.py and hermes_constants.py,
and the ACP Registry manifest/version-lockstep release logic.
- Adds /nix/store/ path detection so `nix run` / `nix profile install`
installs (which don't set HERMES_MANAGED) are correctly identified as
"nix" rather than falling through to "git"/"unknown".
- Retired install-method values ("pip", "homebrew") in existing
.install_method stamps (both code-scoped and home-scoped) are ignored by
the allowlist reader and fall through to "unknown" instead of resurrecting
a retired enum value.
- Updates Nix packaging to ship bare runtime data (locales, optional-mcps)
through store symlinks and wrapper env vars instead of wheel data-files.
- Removes the ACP Registry manifest/icon and their version-lockstep tests.
- Deletes or rewrites packaging, pip-update, Homebrew, and ACP Registry
tests; adds parametrized coverage for the packaging build guard covering
BOTH sdist and wheel paths (the guards live in separate cmdclass entries
— a passing sdist test proves nothing about the wheel path).
- Updates installation/platform documentation and related user-facing copy.
- Adjusts the supply-chain scan so deleted install-hook files do not trigger
a finding, while additions or modifications still require the existing
ci-reviewed label gate.
Supported installation paths (unchanged):
- git installer (install.sh)
- Docker
- Nix/NixOS
- editable development installs (uv sync, uv pip install -e ., pip install -e .)
* fix(desktop): route /compress through session.compress RPC with transcript replacement
Salvages #44462, #53755, and #68218 into a single canonical fix for the
desktop /compress cluster.
The desktop routed /compress through slash.exec, which sends it to the
_SlashWorker subprocess. Compressing a large session outlives both the
desktop's 30s WS timeout and the worker's 45s pipe timeout — the client
gives up, runExec's blanket catch swallows the error, and command.dispatch
surfaces a misleading "not a quick/plugin/skill command: compress" (#44456).
Even when compression succeeded via the _mirror_slash_side_effects path,
the desktop never received the post-compress message list, so summarized
bubbles stayed on screen forever — /compress looked like a no-op.
This change routes /compress to the dedicated session.compress RPC (the TUI's
path), combining the best of all three PRs:
- 120s client timeout matching the TUI's HERMES_TUI_RPC_TIMEOUT_MS (#44462)
- Transcript replacement from the response `messages` via toChatMessages,
the same converter session.resume uses (#68218, teknium1 review on #44462)
- Session-isolation guard: updateSessionState only publishes for the active
runtime, so a late result after a session switch can't clobber the
foreground transcript (#53755, teknium1 review on #53755)
- Coalescing: dedup concurrent compress requests per session (#53755)
- Progress toast ("compressing context...") outside the transcript (#53755)
- Error unmasking in runExec: when slash.exec fails and command.dispatch only
adds "not a quick/plugin/skill command" routing noise, surface the original
worker error instead (#44462)
- /compact alias + focus_topic forwarding
Co-authored-by: AlliDev <AIalliAI@users.noreply.github.com>
Co-authored-by: PinkEVO <PINKIIILQWQ@users.noreply.github.com>
* feat(desktop): route slash commands with dedicated RPCs to those RPCs
Salvages #63513 — introduces a new `rpc` kind on DesktopCommandSurface so
commands with a first-class gateway @method handler bypass slash.exec /
command.dispatch entirely, and a `renderRpcResult` utility that shapes
each RPC's structured reply into readable transcript text.
Migrates 6 commands from exec() to rpc(...):
/agents → agents.list
/save → session.save
/status → session.status
/steer → session.steer
/stop → process.stop
/usage → session.usage
/compress stays as action('compress') — it needs transcript replacement
from the response `messages`, which the generic rpc path can't do (per
teknium1 review on #44462/#63513).
Also includes the json-rpc-gateway timeout message improvement: the error
now includes the configured timeout duration ("request timed out after 120s:
session.compress") so a user can tell whether the default 30s fired or a
per-call override.
Co-authored-by: Jelvin <SmallNew2003@users.noreply.github.com>
* fix(desktop): preserve provider choice during config initialization
* fix(desktop): preserve slash command and host compression semantics
Keep commands whose CLI behavior exceeds their current RPC contracts on slash.exec.
Propagate the full compression timeout through compute-host control, return structured
host compression outcomes with metadata, and retain successful compression feedback
in the desktop transcript.
Add regressions for timeout forwarding, host aborts and metadata sync, structured host
control responses, command routing parity, and numeric stop counts.
* fix(desktop): harden compression state handling
Preserve the invoking stored-session binding for delayed compression results,
normalize replacement histories, and serialize provider selection. Stabilize
gateway platform tests and guard the desktop Git facade during renderer teardown.
---------
Co-authored-by: AlliDev <AIalliAI@users.noreply.github.com>
Co-authored-by: PinkEVO <PINKIIILQWQ@users.noreply.github.com>
Co-authored-by: Jelvin <SmallNew2003@users.noreply.github.com>
* chore(gitignore): ignore installer .install_method stamp
Salvage of #54855 by @drissman — rebased onto current main with root-scoped
rule and sister-marker comments alongside .update-incomplete.
Closes#66189
Root cause: scripts/install.sh writes <install>/.install_method but git
did not ignore it, so managed checkouts show ?? .install_method and
hermes update may autostash the untracked marker.
Fix: add /.install_method to .gitignore (repo-root only).
Verification: git check-ignore -v .install_method
* test(update): assert .install_method survives update autostash (#66189)
Add hermetic regression mirroring the .hermes-bootstrap-complete test:
adopt the real .gitignore, drop the installer .install_method stamp, run
the exact 'git stash push --include-untracked' the updater uses, and assert
the marker is neither swept nor reported dirty. Requested by hermes-sweeper
review on #67364.
Real-world failure from dogfooding the live-theme flow: display.skin was
already 'synthwave' in config, but the desktop never visibly applied it (the
activation event predated the WS transport fix / the connect). The desktop's
gateway.ready seed records the baseline WITHOUT painting (by design — never
stomp the persisted desktop theme on connect), so it believed it was synced.
Re-running 'hermes config set display.skin synthwave' then did nothing twice
over: the watcher signature (name, skin-file mtime) hadn't moved, so no
skin.changed fired; and even on an event, the desktop's name-equality guard
blocked the apply against the seeded baseline.
Two halves:
- hermes_cli: setting display.skin touches the named skin file so the
watcher signature always moves on an explicit set — a same-name re-affirm
now broadcasts skin.changed like any real move. Built-ins (no file) are
unaffected; a name switch already moves their signature.
- desktop: track whether the synced baseline was actually APPLIED vs merely
seeded at connect. A skin.changed matching a seed-only baseline is an
intentional apply and repaints; once applied, repeat same-name events stay
no-ops (protects a manual desktop-side theme switch from snap-back, incl.
across a reconnect re-seed).
When Codex returns 429 usage_limit_reached, Hermes persists the provider's
reset_at on the pool entry and freezes the credential until it elapses --
which can be days out for weekly windows. But the upstream window can
reopen EARLY: the user redeems a banked rate-limit reset (Codex CLI /
ChatGPT UI), upgrades their plan, or OpenAI resets the window. Hermes
never re-checked, so it kept erroring with 'Codex provider quota
exhausted (429); retry after Ns' until a manual re-auth rewrote the
tokens (issue #43747, externally-reset variant).
- hermes_cli/auth.py: add _probe_codex_quota_restored() -- a throttled
(5 min/token) GET of the Codex /usage endpoint; quota counts as
restored when every reported window is <100% used. Add
clear_codex_pool_quota_cooldowns() to lift 429/quota-shaped cooldowns
from persisted pool entries (DEAD and auth-shaped entries untouched).
- resolve_codex_runtime_credentials(): before surfacing a pool-only
cooldown as 'quota exhausted', probe upstream; on a positive probe
clear the cooldown and return the pool credential.
- agent/credential_pool.py: _available_entries() probes frozen
openai-codex entries (clear_expired path only) and unfreezes them when
upstream confirms the reset.
- agent/account_usage.py: a successful /usage reset redemption now
clears persisted pool cooldowns immediately.
Negative paths preserved: probe 429/exhausted/indeterminate keeps the
cooldown; read-only enumeration never probes; non-JWT tokens never
probe (no network in hermetic tests).
The cross-surface theme SDK's live-repaint relies on a gateway skin watcher
that polls config and emits skin.changed on any move. But that emit is
session-less and fires from a background thread, so write_json fell through
its (session-transport -> contextvar -> stdio) ladder to the module stdio
transport — which only reaches the stdio TUI (tee'd to the dashboard WS
publisher). WS clients (the desktop app, dashboard chat) never got it, so
'Hermes themes itself' repainted the CLI/TUI but not the GUI.
Add a live-transport registry (one entry per connected WS peer, maintained by
handle_ws) and a _broadcast_global_event primitive that fans session-less
announcements out to every connected client, falling back to write_json when
none are registered (stdio path unchanged). Route both skin.changed emits
(watcher + the /skin RPC) through it, so a skin switch from any surface
repaints all of them.
Backend-only; desktop already handles skin.changed and does not drop
session-less events.
Two narrow timing windows (reported by null-runner) silently downgraded a
mid-turn correction to a plain next-turn message on the desktop client:
- Turn-build window: a fresh turn flips running=True and builds the agent
asynchronously, so session["agent"] is briefly None. session.redirect
answered 4010 "unsupported", which the renderer's catch swallowed into a
lost follow-up. Queue the correction server-side instead and return
status="queued" — lossless, and honest about what happened.
- Stale runtime id after reconnect: session.redirect 404s on a sid the
gateway no longer maps. redirectPrompt now resumes the stored session and
retries once, mirroring stopPrompt, so a correction fired right after a
reconnect isn't dropped.
The desktop treats "queued" like "redirected": the correction reaches the
model either way, so it's recorded once as a real user message.
The zero-result fallback prefers messages_fts_cjk when built: exact
ranked token match for Latin runs unicode61 fused onto CJK, including
<3-char tokens the trigram leg can't recover.
A pure-Latin query (no CJK characters) routes to the unicode61
`messages_fts` table, whose tokenizer does not insert a boundary between
Latin letters and adjacent CJK characters. Content like "修改youer服务端" is
indexed as a single token, so `search_messages("youer")` returned zero
results even though the substring is present, and the Latin path had no
fallback.
Add a zero-result trigram fallback to the pure-Latin path: when the
unicode61 search misses, retry against the existing `messages_fts_trigram`
table, which matches substrings regardless of word boundaries. The fallback
is gated on `_trigram_available` and on every token being >=3 chars (the
trigram minimum), and only fires on a zero-result miss, so successful Latin
searches keep their unicode61 ranking unchanged.
The trigram query construction shared with the CJK path is extracted into a
`_run_trigram_search()` helper; the CJK branch is refactored to use it with
no behavior change.
Adds regression tests in tests/test_hermes_state.py::TestCJKSearchFallback.
Extract the open_preview emitter into a shared tools/desktop_ui bridge
(one gateway-injected sink, routed by HERMES_UI_SESSION_ID) and add a
second desktop-gated tool on top of it:
- focus_pane(chat|files|terminal|review|sessions) -> pane.reveal event.
The desktop runs each pane's own reveal path (revealDesktopPane table)
and only acts on the active window -- a background turn never moves the
user's focus (desktop AGENTS.md: offer, don't hijack).
open_preview now emits through the same bridge. Both tools are check_fn
on HERMES_DESKTOP (zero footprint elsewhere), sitting beside
read_terminal/close_terminal in _HERMES_CORE_TOOLS.
Deliberately not adding run_slash: letting the agent fire slash commands
mid-turn (/model, /new, /clear) fights prompt-cache + conversation
invariants.
The default `busy_input_mode: interrupt` now redirects the live turn instead
of hard-stopping it and re-queuing a fresh turn, wired consistently across
every first-party surface via the shared core primitive.
- CLI, gateway (busy + PRIORITY paths), TUI (`_handle_busy_submit`), desktop
(`session.redirect` RPC), and ACP call `redirect()` when the agent advertises
`_supports_active_turn_redirect`, and fall back to the proven interrupt +
next-turn queue for older runtimes.
- Redirect is gated to plain text with no attachments: captioned or
attachment-bearing events (including adapters that classify unknown media as
`TEXT`) stay queued so media is never dropped.
- ACP `cancel()` records the interrupted prompt, sets its cancel event, and
hard-stops the agent while holding `runtime_lock`, closing the
cancel-then-correct ordering gap; connection I/O happens after the lock is
released.
- Desktop appends the correction as a real user transcript message so the live
view matches the durable history after reload.
- `/busy` help, onboarding hints, and the new `session.redirect` RPC describe
the redirect behavior; `/stop` remains the hard stop.
Add a desktop-gated open_preview tool so 'open cnn.com in the preview
pane' works. The tool (check_fn on HERMES_DESKTOP, zero footprint
elsewhere) emits a preview.open event through a gateway-injected emitter,
mirroring the close_terminal -> terminal.close bridge. The desktop
handles it in usePreviewRouting, normalizing the target and opening the
pane for the active session only -- a background turn never hijacks it.
Bare domains and localhost are coaxed into fetchable URLs (www.cnn.com ->
https://, localhost:3000 -> http://); file paths and schemes pass through
to the renderer's normalizer.
The Codex app-server runtime bypasses the main conversation loop and drives
its own subprocess turn, so it needs first-class hooks rather than the
OpenAI-loop interrupt path.
- `AIAgent.interrupt()` now forwards a hard stop to
`CodexAppServerSession.request_interrupt()`, and `redirect()` uses Codex's
native `turn/steer` protocol instead of cancelling the subprocess.
- `run_turn()` no longer clears an interrupt that arrived during
`ensure_started()`: a stop landing mid-startup is honored before `turn/start`,
and the interrupt event is cleared on every exit path.
- `run_codex_app_server_turn()` mirrors the loop finalizer's interrupt handoff
(surface `interrupted` / `interrupt_message`, then `clear_interrupt()`) on
both the normal and exception early-return paths, so a hard stop can't leave
`_interrupt_requested` stale for the next turn.
A follow-up sent while the model is still generating previously ended the
turn: Hermes kept only the visible partial text (reasoning was display-only),
cleared the loop, and replayed the message as a fresh next turn. If the
correction referred to something that only appeared in the thinking stream,
the model no longer had that context.
Add `AIAgent.redirect(text)`: a corrective interrupt distinct from a hard
stop. It cancels only the in-flight model request (not tool workers or child
agents), stashes the correction under a lock shared with `interrupt()` so a
concurrent `/stop` always wins, and lets the loop rebuild the same logical
iteration. `_apply_active_turn_redirect()` checkpoints the reasoning that was
actually shown to the user plus any visible partial text as an ordinary
assistant message, then appends the correction as a real user turn — never
replaying incomplete signed/encrypted provider reasoning, and keeping strict
role alternation and prompt-cache stability intact. During tool execution it
degrades to `steer()` so a running tool finishes at a safe boundary.
`_fire_reasoning_delta` now only records reasoning that a display callback
actually consumed, so `show_reasoning: false` never leaks hidden provider
thinking into the persisted transcript.
Follow-up for the salvaged #55800 idle-compaction commit:
- turn_context.py: treat a skipped _compress_context (per-session
compression lock held by another path, failure cooldown, anti-thrash
breaker, codex-native routing) as a strict no-op — only re-baseline
conversation_history and re-anchor current_turn_user_idx after a REAL
compaction. Also re-anchor the user-message index after idle compaction
(the PR predates the reanchor helper).
- hermes_cli/config.py: add idle_compact_after_seconds: 0 to
DEFAULT_CONFIG's compression block (the PR only had
cli-config.yaml.example).
- gateway/run.py: add the idle-compaction status wording to
_TELEGRAM_NOISY_STATUS_RE so the new 💤 message stays out of
human-facing chat surfaces (routine compaction is silent by design);
pin it in tests/gateway/test_telegram_noise_filter.py.
- docs: idle_compact_after_seconds in user-guide/configuration.md and
developer-guide/context-compression-and-caching.md parameter table.
- tests/agent/test_idle_compaction_lock_and_guards.py: end-to-end
coverage with a real AIAgent + SessionDB proving the idle path honors
the per-session compression lock (added after the PR), the persisted
failure cooldown, and the anti-thrash breaker, and that the lock is
released after an idle-triggered compaction.
Salvaged from #55800 by @iso2kx. Implements #27579.
Long-lived sessions (e.g. a Telegram thread resumed over hours/days)
accumulate a large context that the existing size-based threshold only
trims once it crosses `threshold × context_window`. Until then every
turn re-reads the full history, which on large-context models can mean
hundreds of K of cache-read tokens per call even across long idle gaps.
Add a time-based trigger that complements (does not replace) the size
threshold: when a session resumes after `compression.idle_compact_after_seconds`
of inactivity, compact the accumulated history up front, before the first
reply. Disabled by default (0), so existing behaviour is unchanged.
The trigger reuses `_last_activity_ts` (the last time the turn loop did
work) to measure the idle gap at turn start, gates the token estimate
behind a cheap gap pre-check, and skips compaction when the context is
already at/below the post-compression target (threshold × target_ratio)
so a short idle thread never pays for a summarization that saves nothing.
It also defers to an active compression-failure cooldown.
The decision is factored into a pure predicate, `_should_idle_compact`,
which is unit-tested without a live agent.
Composing #57835's multi-fossil summary scan with #47274's merged-handoff
unwrap: when the restart-decay path pulls a merged handoff into the
compression window, its genuine prior-tail user content must enter the
summarizer input (folded into the fresh summary) rather than being
dropped with the summary row. Standalone handoffs still drop. The
continuity test now pins the composed contract: recovered verbatim OR
via summarizer input, never silently deleted, never duplicated.
- Three '_previous_summary == "fresh summary"' exact pins and one
transcript-wide fossil-absence pin predated main's deterministic
task-snapshot grounding (761a0b124e), which prepends a
'## Historical Task Snapshot' section to stored summaries and may
quote a folded head turn inside the handoff. Re-pin the contracts
(fresh body present, fossil absent from non-summary messages)
instead of exact strings.
- Add test_restart_simulation_fresh_compressor_does_not_reprotect_head:
a fresh ContextCompressor over a transcript containing a persisted
handoff summary computes the same decayed protected-head boundary
(compress_start base) as a live already-compacted process, and the
first post-restart compaction does not preserve pre-restart head
fossils (#57814).
protect_first_n decay state (compression_count / _previous_summary) is
in-memory only, so a gateway restart re-protected the persisted handoff
summary and head fossils, growing the head unboundedly across
restart+compaction cycles (#57814).
_effective_protect_first_n now probes a bounded resumed-head window for
a persisted handoff summary (by metadata or content prefix) and decays
protection when one is found, before compress_start is computed. The
first post-restart compaction self-heals: stacked summary fossils are
folded into the next summary prompt instead of preserved verbatim,
rehydration is rolled back on abort, deterministic fallback carries a
bounded redacted previous-summary snapshot, and forced user-leading
merged summaries keep the live tail request after the summary end
marker so they stay rehydratable.
Squashed reapply of the 12-commit series from PR #57835 onto current
main (branch was 1210 commits behind; single add/add conflict with the
task-snapshot grounding helpers resolved by keeping both).
Fixes#57814.
- aux summary call on main intentionally omits max_tokens; use .get() in the
telemetry hook (and widen the param type) so the hook never breaks the call
- update test expectation: aux_output_reservation is None on main
- record no_progress failure_class in the no-progress boundary branch
Follow-up for salvaged PR #60444.
Follow-up for salvaged #24279:
- cli-config.yaml.example: document compression.threshold_tokens
(commented-out, default null = disabled)
- contributors/emails: map maly.dan@gmail.com -> DanielMaly
- tests: should_compress() fires at the absolute cap below the pct
threshold (first-fires-wins); DEFAULT_CONFIG ships None and 0/None
are behavior-neutral incl. across update_model(); the small-context
pct floor is unaffected by the cap and re-derives correctly on
model switch
Add compression.threshold_tokens config option that sets an absolute
token cap for auto-compaction. When configured alongside the existing
ratio-based threshold, the effective trigger point is the lower of the
two, so compression never fires later than the user's preferred token
count regardless of which model is active.
This solves the problem where switching between models with different
context windows (e.g. 1M → 400K) shifts the absolute trigger point,
causing premature or delayed compression.
Rework from PR #24279 addressing sweeper feedback:
- The cap is now a first-class compressor configuration value
(threshold_tokens_cap parameter on ContextCompressor.__init__),
not a post-construction patch on the live instance.
- Applied in both __init__ and update_model() so it survives model
switches and fallback activations (the old approach was undone by
update_model() restoring _configured_threshold_percent).
- Clamped to the model's context length so a cap above the window is
a no-op (ratio-based threshold wins).
- Works with max_tokens output-token reservations.
- Added 9 tests covering cap-vs-ratio selection, model switch survival,
context-length clamping, max_tokens interaction, and invalid values.
- Updated user-facing configuration docs.
- Removed unrelated background-review/curator/Honcho changes (main
already contains background-review memory isolation in 973f27e95).
Config example:
compression:
threshold: 0.50
threshold_tokens: 200000 # never compress later than 200K tokens
Compaction summaries persist across sessions and re-enter every subsequent
summarizer prompt, but every redact_sensitive_text() call in
context_compressor.py used default mode: a no-op under
security.redact_secrets:false, and opaque OAuth-callback / URL-userinfo
credentials passed through even when enabled. The stored _previous_summary
also re-entered the iterative-update prompt unredacted.
Add _redact_compaction_text() — redact_sensitive_text(force=True,
redact_url_credentials=True) — and thread it through all compaction text
boundaries: serializer input (content + tool args), deterministic fallback
summary, summarizer LLM output, manual + auto focus topics, the latest-user
task snapshot, and _previous_summary re-entry.
Note: force=True at this boundary intentionally overrides
security.redact_secrets:false — that opt-out targets live tool output, not
persisted summaries.
Salvages the compaction half of #49556 (the redact.py strict-URL half
landed independently via 75af6dc57/62a00a739). Addresses #43666 item 2.
Co-authored-by: AndrewMoryakov <topazd2@gmail.com>
Integration layer for the cjk_unicode61 tokenizer, rebuilt on the v23
schema (the contributed integration in PR #65544 predated it):
- messages_fts_cjk: external-content FTS5 over a tool-row-excluding view
(same v23 storage discipline as the trigram index it supersedes — zero
inline text copies). Serves EVERY CJK query shape the legacy routing
split between trigram (>=3 chars/token) and LIKE full scans (1-2 char
tokens). Lone 1-char CJK runs and role_filter=['tool'] queries keep
their legacy routes.
- Dedicated marker pair (fts_cjk_rebuild_high_water/progress) gates the
id-scoped triggers, so a cjk-only backfill never gates the complete
messages_fts/trigram triggers.
- Transitions ride (the existing
throttled/resumable chunk engine): fresh DBs are born with the index;
legacy v22 DBs land on v23+cjk in one run; already-optimized v23 DBs
gaining the tokenizer get a marker-gated backfill; live writes are
indexed immediately in every case.
- Tokenizer-loss self-heal: a process that can't load the extension drops
the cjk triggers (writes keep working), leaves a stale breadcrumb, and
the index is rebuilt from scratch on the next optimize run — triggers
are never reinstalled over a gap (external-content 'delete' on an
unindexed rowid is the FTS5 corruption hazard the marker gating exists
to prevent).
- Capability classification: 'no such tokenizer: cjk_unicode61' joins the
degraded-runtime error class everywhere (read probe, write probe,
repair) so tokenizer absence is never misclassified as corruption.
- Config: sessions.cjk_fts (default on, inert without the .so) and
sessions.search_slow_ms in config.yaml, bridged to env by CLI + gateway
(startup + per-turn reload). build.sh falls back to vendored SQLite
headers so no libsqlite3-dev is needed.
Slow-query log path attribution updated: fts_cjk / fts5 / trigram /
like_scan. Tests: 14 lifecycle tests (fresh/legacy/stale/backfill paths,
tokenizer-loss round-trip) + 5 config-bridge tests + slow-log suite.
One INFO line per slow search naming the path taken (fts_cjk / fts5 /
trigram / like_scan), elapsed time, row count, and the query. The 2026-07
session_search investigation needed turn-trace archaeology plus workload
replay to discover that short-CJK queries were full-scanning the table —
with this line the next routing regression is a journalctl grep.
Threshold: sessions.search_slow_ms (default 1000ms; 0 logs every call),
bridged to HERMES_SEARCH_SLOW_MS.
Salvaged from PR #65544 (adapted to the v23 schema in follow-up commits).
Slack delivers user mentions as opaque IDs (<@U123>). The agent had no
way to tell one participant from another — or from itself — so it could
misread a mention of a human as a self-mention and answer messages
addressed to that person (the "bot thinks it's @someone-else" bug).
Two cooperating fixes:
- _humanize_user_mentions rewrites remaining <@UID> tokens (the bot's
own mention is stripped earlier) to @DisplayName in the trigger text
and reply_to_text — the Slack equivalent of Discord's clean_content.
Handles the labelled <@UID|handle> form; unresolvable IDs fall back
to the raw ID.
- _build_identity_prompt injects an ephemeral per-turn system-prompt
line via the channel_prompt seam (applied at API-call time, never
persisted — prompt caching preserved) naming the bot's own workspace
handle (per-team in multi-workspace installs) so the agent has a
positive "that's me" anchor.
Salvaged from #55340 by @benbarclay, rebased over the workspace-scoped
user-name cache (team_id-aware resolution) on main.
In shared Slack threads the model saw only [sender name] prefixes, with
no verifiable current-author Slack user ID — so 'mention me again'
requests could bind to a stale or unrelated <@U...> pulled from names,
memory, or prior history (#17916).
Two cooperating changes:
- The shared-session sender prefix on Slack now carries the current
author's ID from the event envelope:
'[Alice | Slack user <@U123>] ...' — per-turn data, so it does not
touch the cached system prompt.
- The Slack platform notes gain a shared-thread instruction to use the
current turn's sender prefix as the only verified mention target and
never guess or reuse historical mentions.
Fixes#17916.
Salvaged from #18711 by @LeonSGP43 (asdigitos), rebased over the
sender-name neutralization added on main (the ID is appended after
neutralizing the display name; the ID itself comes from the Slack
event, not user-editable text).
Two mention-tracking gaps around thread parents (#24848):
1. When a thread PARENT @-mentioned the bot (e.g. '<@bot> check this
and ask me before running'), a later bare reply like 'run' fell
through every wake check if the mention event predated this process
(restart) — _mentioned_threads is in-memory only. Add a 5th wake
check that fetches the parent text (with the bot mention preserved
via strip_bot_mention=False) and wakes when the parent addressed the
bot, registering the thread so later replies skip the fetch.
2. A TOP-LEVEL @mention starts a thread (session keying falls back to
the message ts), but only the raw event thread_ts was registered in
_mentioned_threads — so replies to a top-level mention did not
auto-trigger. Register the session-scoped thread_ts instead.
_fetch_thread_parent_text reuses the shared thread-context cache (raw
payloads) so the parent check costs at most one conversations.replies
call per thread; _register_mentioned_thread centralizes the bounded-set
eviction.
Salvaged from #24848 by @kaiyisg, rebased onto the extracted
_should_wake_on_unmentioned_message helper.
The un-mentioned wake decision relied on three checks: thread root in
_bot_message_ts (populated only by the adapter's own send() path),
_mentioned_threads (populated on @mention), and an existing session.
Two gaps (#63530):
- Gap A: bot messages posted OUTSIDE gateway send() — skills/scripts
calling chat.postMessage directly, cron/API posts — never enter
_bot_message_ts, so human replies in those threads were silently
dropped.
- Gap B: _bot_message_ts is process memory; after a gateway restart the
bot stopped waking on replies to threads it started before the
restart.
Fix: add a 4th, API-derived check — _bot_authored_thread_root — which
resolves the thread root's author via conversations.replies (cached in
_thread_context_cache via the new parent_user_id field, TTL-bounded).
Root authorship comes from Slack itself, so it covers outside-send
posts and survives restarts, unlike in-memory ts tracking. The wake
decision is extracted into _should_wake_on_unmentioned_message for
direct unit testing; the legacy checks remain first (cheap, additive).
Fixes#63530.
Salvaged from #64067 by @knoal (author metadata normalized to their
GitHub identity), rebased over the thread-context formatter split and
extended with per-team bot-id resolution for multi-workspace installs.
Persistent sessions survive gateway restarts, but thread replies posted
while the gateway was DOWN never reached the session — and the adapter
had no way to notice, so the conversation silently resumed with a hole
in it.
On the first ordinary reply per thread after a restart (tracked by a
fresh-process _thread_rehydration_checked set), fetch the thread delta
past the persisted per-session watermark and inject any missed messages
as part of the new turn via channel_context. Exactly-once per thread
per process; when the watermark is empty (pre-feature sessions) the
check is a no-op. Steady-state replies keep advancing the watermark so
rehydration never re-injects messages the session already carries as
ordinary turns. Prior history is never rewritten (prompt caching safe).
Builds on the persisted watermark introduced for #23918.
Salvaged from #33215 by @vexclawx31, reworked from a repeated
full-thread injection guard into a watermark-delta injection so
rehydration adds only what the session actually missed.
Once a thread has an active session, a later reply that explicitly
@mentions the bot did not re-fetch Slack thread context, so the agent
missed messages added to the thread after the initial hydrate (e.g.
other bots/integrations replying in multi-agent workflows). The
explicit mention is a fresh intent signal and now triggers a refresh.
Mechanics:
- SessionEntry gains a small persisted metadata dict, with
SessionStore.get/set_session_metadata accessors (survives gateway
restarts via the routing index).
- The adapter stores a per-thread consumption watermark
(slack_thread_watermark:<channel>:<thread>) recording the last
thread ts the session consumed.
- On explicit mention in an active thread, _fetch_thread_context runs
with force_refresh=True (bypassing the TTL cache) and after_ts=<the
watermark>, so only NOT-yet-seen messages are injected — as part of
the new turn via channel_context. Prior conversation history is
never rewritten, preserving prompt caching.
- _fetch_thread_context caches raw conversations.replies payloads so
watermark-scoped re-formatting needs no extra API call; formatting
is split into _format_thread_context.
- Thread session keys are built once in _build_thread_session_key
(shared by the wake gate and the watermark accessors), still via
build_session_key().
Fixes#23918. Supersedes #62299 (keyword-triggered refresh limited to
'investigate' prompts — the mention signal is the general fix).
Salvaged from #23927 by @heathley, rebased onto the plugin adapter
layout and rerouted through channel_context instead of text-prepend.
Prepending cold-start thread backfill directly onto the message text
moved a recognized command (e.g. a bang-normalized "!queue ...") away
from character zero, so downstream command routing misclassified it as
conversational text and the command silently didn't run.
Route the backfill through MessageEvent.channel_context instead —
gateway.run already prepends channel_context after command dispatch
("[New message]" framing), so commands keep their COMMAND type while
the recovered history stays available to the agent.
Supersedes #68020, which dropped the fetched context entirely for
commands instead of preserving it out-of-band.
Salvaged from #66069 by @LevSky22.
