The frost is native vibrancy, which is the window's content view rather than
an element — it fills the whole rectangle and nothing in the page can clip it
to the sheet. That was only ever right while the sheet covered the window;
anywhere it falls short, the difference is frost over empty space. On a fresh
thread the sheet is zero and the difference is the entire window, which is the
grey slab that appears the moment you put the caret in the composer.
Gating the caller's `engaged` was not enough, and is why the first attempt at
this missed: the hook turns the frost on for a focused composer by itself,
independent of what the caller passes. The veto belongs inside, next to that
check.
The bundled docx, xlsx, powerpoint, and pdf skills were adapted from
Anthropic's document skills and carried their proprietary LICENSE.txt
(no derivatives, no redistribution). Flagged as critical license
findings by the SkillEvaluator Tier 1 scan of our skill tree.
This replaces all four with clean-room rewrites:
- Authored from scratch against library knowledge only (python-docx,
openpyxl, python-pptx, pypdf/reportlab/pdfplumber — all MIT/BSD) by
isolated subagents given functional specs, with an explicit
prohibition on reading the prior skill content or anthropics/skills;
session transcripts retained as provenance evidence.
- MIT licensed (LICENSE file per skill), author: Nous Research.
- Each skill: SKILL.md to house standards + argparse helper scripts
with UTF-8-explicit I/O + its own e2e pytest suite (fixtures built
on the fly, non-ASCII round-trips run under LC_ALL=C).
- All four pass SkillEvaluator Tier 1 pii+unicode+lint 3/3.
tests/skills/test_office_document_skills.py rewritten against the new
contracts: MIT/no-Anthropic-text invariants, scripts documented in
SKILL.md, argparse CLI shape, and a no-locale-default-open() check
(which caught and fixed a real gap: pdfplumber text reads are fine,
but the invariant scan now guards every future script).
Docs pages regenerated for the four skills (scoped; unrelated
generator drift excluded).
Honest capability deltas vs the old versions are documented per
SKILL.md (e.g. tracked-changes accept/reject and OOXML XSD validation
are not reimplemented; form flattening limits stated).
Findings from scanning skills/ + optional-skills/ with NVIDIA
SkillEvaluator's deterministic Tier 1 checks (PII/secrets, unicode
smuggling, script lint):
- pixel-art, pokemon-player: remove hardcoded /home/teknium/ personal
paths (use ~ / portable phrasing); pokemon-player no longer claims
machine-specific state as fact
- kanban-video-orchestrator: replace <path> angle-bracket token in
frontmatter credits (flagged as XML-in-frontmatter prompt injection)
- comfyui, hermes-agent, unsloth, 1password, actual-setup: rephrase
placeholder secrets so they no longer pattern-match real credentials
(your-* placeholder convention, comment markers, {env:...} form)
- docker-management, pytorch-lightning: drop user:pass@ from example
connection strings (env/secret-manager guidance instead)
- evm: break up Keccak round constant that Luhn-validates as a credit
card number (digit-group underscores, value unchanged)
All targeted skills now pass pii+unicode+lint 3/3 except unsloth, which
retains scanner false positives only (Colab notebook IDs read as Bitcoin
addresses; an email inside a quoted upstream system prompt).
Fixes 13 issues found in PR #20774 review:
1. Wiring: engine selection moved from run_agent.py to agent/agent_init.py
(where init_agent lives on current main). Transform hook moved from
run_agent.py to agent/conversation_loop.py (where run_conversation lives).
2. Prompt caching: replace copy.deepcopy with copy-on-write (shallow list
copy + clone only messages that are mutated). Use last_prompt_tokens
from update_from_response instead of re-estimating tokens every call.
System extension injection is idempotent (one-time cache break).
3. Signature mismatch: _message_signature renamed to _content_signature
and now excludes tool_calls/tool_call_id from the hash. This prevents
mismatches when _canonicalize_api_tool_calls re-serializes argument
JSON with sort_keys=True on the API copy.
4. update_model: accepts api_mode parameter (required by agent_init.py).
5. Reconciled with select_context: transform_api_messages is a separate
hook that runs AFTER select_context and sanitization, before
prompt-cache marker placement. Both hooks coexist with clear ordering.
6. Dedup/purge: kept as DCP-specific strategies (different semantics from
ContextCompressor._prune_old_tool_results — DCP deduplicates by
tool+args signature, not by content hash).
7. Removed copy.deepcopy: replaced with shallow list copy + copy-on-write
via _clone_if_needed. Only messages that are actually mutated get
cloned.
8. Removed redundant _ensure_refs call: _match_api_messages_to_refs no
longer calls _ensure_refs (the caller already called it).
9. _message_key still uses index (needed for positional ref assignment),
but _content_signature is cached per id(msg) to avoid re-hashing.
10. _inject_nudge: only injects into user messages, never falls back to
non-user messages (prevents role semantics violations).
11. Memory: _evict_inactive_blocks bounds blocks_by_id to
_MAX_INACTIVE_BLOCKS (50) deactivated blocks.
12. Merged _range_tool_schema and _message_tool_schema into a single
_compress_tool_schema. Merged _handle_range_compress and
_handle_message_compress into _handle_compress.
13. Dropped DCP_CONTEXT_ENGINE_PR_SPEC.md (temporary file, not for tree).
Config defaults kept minimal in hermes_cli/config_defaults.py (only
the keys the engine actually reads, not the full DCP-compatible surface).
Closes#20717
Follow-ups on the salvaged commit (#37207 by @charzhou):
- Persistent /root home mount translates too: an agent writing
/root/out.png produced a real host file under
<sandbox>/docker/default/home the gateway could not find.
- /root/.hermes cache mounts translate to the HOST cache (longest-prefix
beats the home mount), so MEDIA:<agent_visible_image> paths deliver.
- /root/.hermes/* OUTSIDE a cache mount never translates through the home
mount: those are the sandbox's credential copies (.env, auth.json) that
sit outside the host-side denylist prefixes — fail closed.
- Run the idempotent terminal-config->env bridge before mount parsing so
in-process gateways (Desktop backend, hermes serve) see the active
backend and docker_volumes (covers #42299's /output case there too).
Translate MEDIA paths under configured Docker volume mounts (and the
default persistent /workspace) to host paths before media delivery
validation, using longest container-prefix match so host:/workspace and
/output export mounts work.
A fresh thread has nothing to show, but the HUD showed a slab of frosted
glass above the bar anyway. Vibrancy is the window's whole content view, so
it frosts the full rectangle — fine while the band always filled the window,
wrong the moment there is no transcript to fill it. It stays off until there
is something to back.
The sheet had a 12px floor for the same reason: the breathing room above the
first row was added in CSS, so a zero-row transcript still measured 12. It is
folded into the measured height now and only applies when there are rows.
Salvage of PR #17973 by @TKCen (Sebastian Hänisch), re-implemented on
current main to preserve speed/instructions/provider params,
prepare_spoken_text normalization, OPUS_VOICE_PLATFORMS, is_write_denied
path security, microsecond timestamps, and the streaming-TTS gate.
- Split long TTS text into provider-safe chunks instead of truncating
- Pack generated audio against platform upload limits (Discord 10MB,
Telegram 50MB, configurable via tts.delivery_profiles)
- Combine chunks with ffmpeg (OGG/Opus re-encoded, MP3 stream-copied)
- Multi-file delivery when combination fails or would exceed limits
- Remove hard [:4000] truncation from all callers (cli.py, voice.py,
gateway/run.py, gateway/platforms/base.py)
- Gemini TTS raises ValueError instead of silently truncating when
composed prompt exceeds the provider limit
Simplify-code fixes: removed dead all_touched_paths set, added
try/finally for scratch file cleanup on exception, clean error response
on chunk failure instead of leaking stale file_path.
The published tarball ships lib/binding/napi-9-darwin-unknown-arm64 on every
platform, so a real Windows host has both it and the downloaded win32 binding
— the classify-everything gate threw on the darwin dir and killed every
Windows pack. Stage only bindings naming the target platform (classify still
rejects impostors), stop copyGlobByExt from recursing into lib/binding, and
add a version tripwire so a get-windows bump fails the build until the
lib/windows.js rewrite is re-verified.
Also from review: the renderer answers window.read.respond with empty text
when the IPC invoke rejects (older shell / main-side throw) instead of
stalling the tool's 30s timeout; the tool schema discloses that sibling
Hermes windows are skipped; docs gain read_window_below in both references.
The desktop_ui and post-hook ownership contract tests enumerate their tool
sets exactly — add read_window_below to both (plus the executor-path
parametrize case). Lint: sorted type import, explicit GetWindowsModule type
instead of an import() annotation, curly + blank-line style.
get-windows@9.3.0 (MIT, zero runtime deps on macOS/Linux) is external to the
esbuild bundle and staged into dist/node_modules per target platform: the
universal Swift helper on macOS, the prebuilt N-API binding on Windows
(fail-closed magic-byte validation), nothing on Linux (xprop at runtime).
The staged lib/windows.js is rewritten to load the binding directly so
@mapbox/node-pre-gyp's tree stays out of the package.
New electron/window-below.ts: pure z-order picker (walks past our own pid,
first other-process window whose bounds overlap ours) over get-windows'
front-to-back enumeration, with the Linux xprop stacking order reversed to
match (EWMH _NET_CLIENT_LIST_STACKING is bottom-to-top). Main answers the
hermes🪟readBelow IPC; on macOS other apps' titles pass through only
when Screen Recording is already granted — never prompted for.
Desktop-gated (desktop_ui toolset) metadata-only window awareness: the agent
can ask which application window sits directly behind the Hermes window
(app, title, bounds — never pixels). Rides the same blocking bridge as
read_terminal: the gateway emits window.read.request and the renderer
answers window.read.respond.
Pages serves exactly the newest artifact, so every push-triggered deploy
deleted the previous build's content-hashed JS/CSS while edge caches
(max-age=300, stale-while-revalidate=3600) kept serving HTML that
referenced them. With deploys landing every ~15-30 min, docs pages spent
most of the day pointing at 404'd bundles — search (pure client JS) was
the loudest casualty.
Fix: keep a rolling 14-day pool of hashed assets (en + zh-Hans) in the
Actions cache and union-merge it into each deploy artifact, current
build authoritative on collision (cp --update=none). Stale HTML and
already-open tabs now keep resolving across any number of deploys.
4c2961c51 added referenced_skill_names() so the curator never archives a
skill a cron job depends on — paused jobs and infrequent schedules would
otherwise age their skills out and the next run fails to load them.
62972060c then taught the scheduler that jobs may store ABSOLUTE skill
paths, normalizing them through normalize_skill_lookup_name before
skill_view. The protection set kept returning the raw string, so it now
holds a full path while the curator matches it against bare skill names.
Those jobs silently lost their protection: the skill is archived, and the
next fire logs a warning and runs the job without its instructions.
Canonicalize each reference the same way the scheduler resolves it, with
a deferred import and a verbatim fallback so a resolver failure can never
drop a name (referenced_skill_names has exactly one caller, the curator's
protection lookup, so nothing else sees the change).
The brief is single-connector (every command comes from google-workspace),
so it ships as references/daily-brief.md with a pointer + load trigger in
SKILL.md — progressive disclosure instead of a new skill-index entry.
Contributor's procedure preserved (half-open day windows, mail-to-meeting
linking with fuzzy-match discipline, 7-section brief, bounded actions);
credit noted in the reference header. Tests (8) guard the wiring and
disciplines. Version 1.1.0 -> 1.2.0.
_humanize_user_mentions rewrites <@UID> to @DisplayName by passing the
resolved name as re.sub's replacement, where re parses it as a template.
A display name is arbitrary user-set text, so the escapes in it are the
user's characters, not regex syntax:
dev\ops -> re.error: bad escape \o
a\1b -> re.error: invalid group reference 1
\g<0> -> expands to the whole match, silently putting the opaque
<@UID> back — the token this method exists to remove
The trigger-text call site sits in _handle_slack_message outside any try,
and both Bolt event handlers await it bare, so the raise takes the whole
inbound message down: every message mentioning that person is dropped.
Pass the replacement as a function instead — re does no template parsing
on the return value, so the name lands verbatim. Same shape the Matrix
adapter already uses for its outbound mention rewrite.
The coverage warning listed bare page ranges, which tells the agent
WHERE the gaps are but not WHAT they contain — its only options were
guessing or OCRing everything. Each gap is now labeled with the last
text extracted before it (usually a section divider page), so the agent
can decide which gaps it actually needs and render/OCR only those.
Gap list capped at 20 entries with a summary line for pathological
alternating documents.
Follow-ups on the salvaged commit:
1. get_cache_directory_mounts() now CREATES missing staging dirs instead of
skipping them. Docker snapshots the mount list at container creation, so
a dir born later (first attachment, first clipboard image) dangled for
the life of a persistent container. Empty bind mount costs nothing.
2. to_agent_visible_cache_path() translates per-backend instead of
docker-only: docker/modal -> /root/.hermes, ssh/daytona/vercel_sandbox ->
~/.hermes (shell-expanded remotely; bytes arrive via file sync), local/
singularity keep the host path (apptainer auto-binds the host home).
Mirrors the proven _agent_cache_base_for_env heuristics.
Updated the two mount-list tests pinning the old skip behavior; added
per-backend translation coverage.
When extraction of a binary document format (.pdf, .docx, .xlsx, Office,
EPUB…) fails for a specific reason — the anydoc size cap, an encrypted or
malformed file — read_file previously swallowed the ExtractionError at
debug level and fell through to the generic 'Cannot read binary file'
guard, so the agent never saw the actionable reason (e.g. 'Document too
large to convert (N bytes, limit is 52,428,800)').
read_file now returns the specific extraction failure for binary document
formats. Fallthrough behavior is preserved where a raw read is still
useful: .ipynb (plain JSON) and converter-unavailable PDFs keep their
historical raw-read path, and the 'Unsupported document type' shape (no
extra information) keeps the generic guard.
Follow-up to #80004, where the size-cap message was being generated but
never reached the agent.
Follow-up on the cherry-picked contribution from @michaelsam94 (#51794):
replace display-string/exact-value snapshot assertions with invariant
checks per the no-change-detector-tests policy. Live-tested
fal-ai/nano-banana-2 and fal-ai/nano-banana-2/edit through the real
payload builders: both pass.
Rewrote from a sibling-skill routing table into a skill that carries its
own procedure, and folded in generalized rules from maintainer practice:
- full-thread reads (gh issue view --comments; newest comment = live state)
- duplicate-PR sweep (issue number + keyword variants) before any code
- design-intent check via git log -p -S alongside premise reproduction
- fix the class: sweep sibling call sites into the same PR
- sabotage run: prove the regression test fails without the fix
- open the PR immediately (PR dispatches CI; CI latency is the long pole)
- close the loop: comment the issue with the PR link
Also: description 205 -> 59 chars, author credits Ben Barclay first,
modern section order, boilerplate trimmed, tests (10) incl. a
router-pattern guard, scoped docs regen.
The non-streaming /v1/responses path built function_call and
function_call_output output items with no status field (and no item id),
while the SSE streaming path correctly emits status in_progress ->
completed. Spec-strict OpenAI clients reading the non-streaming output
array could interpret the status-less function_call items as pending
calls the CLIENT must execute — but these tools were already executed
server-side by the Hermes agent and are replayed for structured tool UI
only. Reported by a community user whose GPT-5.6 client concluded 'a
server should not tell an OpenAI client to execute a tool the server
already executed itself'.
- _extract_output_items now stamps status: completed and spec-shaped
item ids (fc_/fco_) on replayed items, matching the streaming path
- test updated to pin status + id shape
- docs example updated + explicit note that output tool calls are
replayed, never pending
Four fix-forwards from the adversarial post-merge audit of the Aug 7
unreviewed merge batch:
- estop (#81148): is_engaged() now fails SAFE (engaged) on stat errors;
the gateway estop gate lets recognized slash commands and replies owned
by in-flight work (update prompts, clarify, slash-confirm, tool
approvals, running sessions) through instead of consuming them; new
gateway /pause [reason|off] command gives messaging-only operators an
in-band engage/resume path (busy_policy=dispatch so it works mid-run).
- cron monitor mode (#81138): execution-mode invariants (monitor x
no_agent, monitor_script x monitor_url, no_agent-requires-script) now
have ONE owner (_validate_job_mode_invariants) called from BOTH
create_job and update_job, so the create-time invariant can no longer
be silently violated through the update door.
- cron notepad (#81139): remove_job now clears the job's notepad rows
(clear_notepad was dead code -> orphaned KV state forever); clear is
best-effort and no-ops without creating notepad.db.
- delegation batch gate (#81141): template-marker regex narrowed to
multi-word placeholder shapes only (<feature name>, {file_path}) so
generics (Vec<T>), HTML tags, JSON snippets, glob braces and f-string
style no longer reject legitimate batches; duplicate-goal rejection
removed (best-of-N fan-outs are legitimate).
The salvaged bytes path (_extract_anydoc_bytes) bypassed the coverage
check added in #81680. Materialize transferred PDF bytes in a host temp
file for the pdftotext scan, and name the backend-visible path in the
recovery command rather than the temp file.
Video (plugins/video_gen/fal): Seedance 2.5, MiniMax H3, Seedance 2.0
Mini, FLUX 3, Grok Imagine 1.5, Gemini Omni Flash (i2v-only). New
family capability flags:
- duration_int: endpoints that take duration as a JSON integer
- resolution_aliases: maps 720p/1080p-style values onto non-standard
enums (H3's 768P/2K/4K)
- image_drop_keys: strips keys the family's i2v endpoint rejects
(aspect_ratio on Seedance 2.5 / H3 / Grok 1.5)
Image (tools/image_generation_tool): Seedream 5.0 Pro (+edit) and
Lite, Ideogram V4 instant + fast, Qwen Image 3 (+edit), MAI Image 2.5
Pro, Nano Banana 2 Lite (+edit), Recraft V4.1.
Every new endpoint live-tested against fal.run through the real
payload builders + submit path: 18/18 pass (t2v, i2v, t2i, and edit
probes). Note: several new endpoints return HTTP 409 from the Nous
Portal FAL proxy allowlist until it is updated portal-side; BYOK
FAL_KEY works today and the existing 4xx guidance message covers it.
Force-redact every terminal exception and traceback field before JSON serialization, including environment creation, background startup, exhausted foreground retries, and the outer catch-all. Preserve the current command-aware, opt-out-respecting redact_terminal_output(output, command) behavior for successful output.
anydoc converts the PDF text layer only and emits no image placeholders
or page markers, so a mostly-scanned PDF extracts 'successfully' into
section headers with empty bodies — silent data loss the model cannot
detect. Count per-page text via poppler pdftotext and prepend an
EXTRACTION COVERAGE WARNING naming the empty pages and the recovery
path (pdftoppm + vision_analyze, or the ocr-and-documents skill).
Found on a 311-page HOA resale package where 198 scanned pages
(CC&Rs, Bylaws, Articles, insurance certs) vanished without a trace.
Brooklyn Nicholson
Teknium
kshitij
Jack Maloney
CharZhou
Ben Barclay
hermes-seaeye[bot]
Drexuxux
fangliquanflq
webtecnica
michaelsam94
Coy Geek