Empty model.base_url plus a runtime custom-provider URL was treated as a
route mismatch, so gateway session-reset banners dropped model.context_length
and fell back to the Qwen family default (131K) while /status still showed
the configured 262K pin.
test_custom_provider_context_avoids_false_shrink_warning used
qwen3.8-max-preview as a slug that deliberately falls through to the
generic 'qwen' 131K catalog match. The new qwen3.8-max
DEFAULT_CONTEXT_LENGTHS entry (1M) now substring-matches the preview
slug too, so the no-custom-providers branch stopped warning. Swap the
fixture to qwen3.9-max-preview, which still hits the generic fallback
— the test's intent (custom_providers threading) is unchanged.
Qwen3.8 Max is live on both OpenRouter and the Nous portal
(qwen/qwen3.8-max, 1M context, 131K max output). Per the
newest-max-replaces-last-max convention, it takes qwen3.7-max's slot
in both curated lists.
- hermes_cli/models.py: OPENROUTER_MODELS + _PROVIDER_MODELS[nous]
swap qwen/qwen3.7-max -> qwen/qwen3.8-max
- agent/model_metadata.py: DEFAULT_CONTEXT_LENGTHS entry for
qwen3.8-max at 1,000,000 (verified against OpenRouter live
metadata and Nous /v1/models 2026-08-03)
- tests/test_empty_model_fallback.py: swap incidental catalog fixture
to the surviving slug
- website/static/api/model-catalog.json: regenerated
Pricing snapshot skipped: both routes bill via official_models_api
(live pricing), verified with resolve_billing_route. Reasoning
timeout floor already covered by the qwen3 prefix (180s).
The reset keywords have existed in both CLI and gateway handlers since
June but were undocumented — users couldn't find how to cancel a
personality overlay. Adds a 'Resetting to the default' section to the
personality feature page and mentions the reset in the CLI guide,
slash-command reference (both tables), and messaging command table.
Cherry-picked from PR #58560 by @itsflownium, adapted to current main
(_getenv instead of os.getenv). Moves ANTHROPIC_API_KEY check ahead of
Claude Code credential file and credential_pool auto-discovery so an
explicitly configured key is never shadowed by auto-discovered OAuth.
Fixes#58546
CI-caught: compute_prompt_breakdown still looked for <available_skills> in the stable tier, but #37117 moved it to volatile. Search volatile first, fall back to stable for older sessions.
Simplify-pass finding: the safety note still cited 'skills edited' as a stable-tier input whose change mismatches the rebuilt prefix — after this PR a skill edit changes only the volatile tail (that's the point). Swap the example for genuinely stable-tier inputs.
The skills index is runtime-mutable: the agent adds and patches skills mid-session, so it is not byte-stable. Keeping it in the stable band breaks that band prefix-cache contract, because every skill edit changes the stable band and invalidates the entire cached prefix in front of it. Move it to the front of the volatile band so the stable scaffold (identity, tool guidance, model guidance) stays cacheable across skill edits.
`nix/devShell.nix` collected `devShellHook` by scanning every package:
nonNpmHooks = map (p: p.passthru.devShellHook or "") packages;
But `minimal` and `messaging` are `.override` variants of `default`, so
each carries its own `devShellHook` exporting its own HERMES_PYTHON. The
scan therefore concatenated three conflicting exports and forced Nix to
evaluate and realise three separate uv2nix editable venvs on every
`nix develop`.
`attrValues` is alphabetical, so the last hook won (`minimal`) while
`python`/VIRTUAL_ENV came from `default`'s devDeps:
HERMES_PYTHON = ...dimim2... (minimal — no optional deps)
python / VIRTUAL_ENV = ...85r28... (full)
Inside the shell `$HERMES_PYTHON -c "import anthropic"` failed while
`python -c "import anthropic"` succeeded. Worse, `scripts/run_tests.sh`
prefers HERMES_PYTHON, so the suite ran against the minimal venv. Its
guard did not catch this: it only checks that HERMES_PYTHON has pytest,
and minimal's venv does (pytest is in the `dev` group), so the wrong
interpreter was silently accepted.
Tying the hook to `packages.default` — the same package whose `devDeps`
are installed — keeps HERMES_PYTHON, `python`, and VIRTUAL_ENV pointing
at one venv by construction.
editable venvs referenced 3 -> 1
their combined closure 421 MB -> 140 MB
test failures 85 -> 32
The venv mismatch was masking 53 failures; e.g. test_web_tools_config.py
goes 2-failed -> 38-passed. Full suite is now 25369 passed / 32 failed,
and those 32 reproduce identically on a pristine HEAD worktree with no
nix/ changes under the same interpreter (mostly NixOS artifacts — tests
spawning bare `python3` in a scrubbed env exit 127).
The PR's concurrency wrapper splits call_llm into a semaphore-guarded
entry + _call_llm_impl; main added extra_headers to call_llm's
signature after the PR's base, so the split has to forward it too
(dropped silently otherwise — Azure Foundry and custom-endpoint
callers set it).
Simplify-pass follow-up on the #70782 salvage: _bearer_auth_headers
hard-emits ${MCP_<NAME>_API_KEY} but install_entry only persists
auth.env-declared vars — a manifest naming its key differently (the
shipped n8n style) would install cleanly yet send a literal-placeholder
header at connect time (silent 401, the #37792 bug class). Enforce the
naming contract at parse time. Also pins the secret-stays-in-.env
property in the install test (raw config.yaml carries the template,
never the secret). Mutation-checked: validation disabled -> guard test
fails.
A successful compaction frees the largest allocation a long session ever
drops (the compressed-away message dicts), but Python's arena allocator
keeps those pages in the heap — RSS retains the pre-compaction
high-water mark until exit. #76905's trim_memory lifecycle covers the
gateway/TUI housekeeping loops but not the CLI compression path.
Call trim_memory(reason='post-compression') at the compression-success
point in ContextCompressor.compress(), following the house pattern
(lazy import in try, debug-level log on failure). The helper is
glibc-gated, config-gated and rate-limited, so it is a safe no-op on
other platforms and cannot fail compression.
Re-expresses the intent of #70782 (JonthanaHanh), which reached for a
bare gc.collect(); trim_memory is the house mechanism and already
wraps a collect.
When an optional-mcps manifest declares transport.type=http with
auth.type=api_key, install_entry() prompts for the key and saves it to
.env, but _build_server_config() only handled the oauth case — the
api_key case produced a bare url entry with no headers, so every
request to the server was unauthenticated (-> 401).
Reuse _bearer_auth_headers(entry.name) from mcp_config.py so the
catalog path emits the same 'Authorization: Bearer ${MCP_..._API_KEY}'
template as the manual 'hermes mcp add --url' path.
Salvaged from #70782 (production hunk applied clean; tests re-anchored
onto current main). Credit: JonthanaHanh.
Review follow-up on the #30223 salvage: the original changed the base
cooldown from 60s to 1800s, benching the primary for 30 minutes on the
FIRST 429 (30x regression in primary-restore latency) and breaking the
existing test_rate_limit_exhaustion_keeps_60s_cooldown contract.
Keep upstream's 60s base and escalate per consecutive rate-limit:
60s -> 2m -> 4m -> 8m -> ... capped at 4h. Counter still resets on
successful primary restore (cicae's mechanism, unchanged).
New tests: escalation doubling, 14400s cap, reset-on-restore.
Existing 60s contract test passes UNCHANGED. Mutation-checked:
escalation disabled -> 2 fail; reset disabled -> 1 fails.
Replace the fixed 60-second cooldown with exponential backoff:
30min → 1h → 2h → 4h cap.
The counter is reset by restore_primary_runtime on successful
primary-provider recovery, so the backoff is strictly for
consecutive failures within a single degradation window.
Closes#29702
`DiscordAdapter.disconnect()` cancelled the bot task before tearing down voice
clients. `leave_voice_channel()` ends in `await vc.disconnect()`, and discord.py
sends a voice state update over the main gateway websocket and then waits for the
voice socket to close. The bot task is the loop running that gateway connection,
so cancelling it first left the handshake with no transport: it could never
complete and blocked until the caller's shutdown timeout fired.
The effect was a fixed ~5s penalty on every shutdown with a voice connection
open, ending in "discord disconnect timed out after 5.0s - forcing continue",
with the voice disconnect abandoned rather than completed.
Measured on a live gateway with a voice connection open in both cases:
before: timed out after 5.0s, all adapters disconnected at +5.29s
after: discord disconnected (0.12s), all adapters disconnected at +0.46s
Moving the voice-cleanup loop above `_cancel_bot_task()` preserves the
zombie-client protection its comment describes: the bot task is still cancelled
before `client.close()`, just after voice teardown rather than before it. Voice
teardown is the one step that still requires a live gateway.
Adds a regression test asserting the ordering. It fails on the previous ordering
at index 1 with `cancel_bot_task != leave_voice_channel:111`.
Fixes#76044
Salvage of #71282 (Fixes#71281): a routable-but-dead endpoint (corp
LAN address while off-VPN) blackholes TCP SYNs, so every probe in the
model-metadata waterfall waits out its full connect timeout — 20+
seconds of stall per startup across detect_local_server_type,
fetch_endpoint_model_metadata, and the per-model probes.
A module-level blackhole cache keyed on host:port is populated when
any probe observes a ConnectTimeout (httpx or requests; read timeouts
deliberately excluded — an accepted connection is not a blackhole) and
consulted at the top of each guarded function. 30s TTL: long enough to
collapse one startup burst, short enough that VPN recovery is picked
up without a restart. Guard ordering: blackhole check -> disk L2 ->
HTTP waterfall, and a blackholed leg aborts the remaining legs instead
of letting each stall in turn.
Squash of the PR's two real commits (the branch's merge commits made
it un-rebase-merge-able; content verified identical via merge-tree).
When a Discord channel message initiates a relay auto-thread, the thread does
not exist at ingest (source.thread_id is None) — the connector creates it on
its FIRST send and auto-threads any outbound carrying the reply anchor. The
final reply carries that anchor, so it lands in the thread. But the
tool-progress / status bubbles (the "Searching the web for..." updates and the
streaming preamble) were sent with _progress_metadata=None and
_progress_reply_to=None: _resolve_progress_thread_id returns None for Discord
(only slack/mattermost get a synthetic thread), so the progress send had no
anchor and the connector posted it FLAT in the parent channel. Result: the
search-status updates leaked outside the thread while the answer threaded
(staging repro 2026-08-02).
The connector now stamps prospective_thread_id on the inbound (the anchor
message id == the id of the thread it will create). Reuse it: when a
relay-delivered Discord channel-initiate carries prospective_thread_id and has
no real thread yet, carry the reply anchor (event_message_id) on both the
progress metadata (reply_to_message_id) and the progress reply_to, so the
connector routes the progress bubble into the SAME auto-thread as the final
reply. Applied to both the tool-progress path (_progress_metadata /
_progress_reply_to) and the status/interim callback path
(_status_thread_metadata). Events already arriving in a real thread, DMs, and
non-relay sources are untouched (guarded on delivered_via_upstream_relay +
prospective_thread_id + not thread_id).
Tests: two new cases in test_run_progress_topics.py — a relay Discord
channel-initiate asserts every progress send carries the anchor (reply_to +
metadata.reply_to_message_id + non_conversational), and an event already in a
real thread asserts the synthetic-anchor path does NOT engage. Full gateway
progress + relay + session suites green (228 passed).
Salvage of #26860 (hunk 2, ported \u2014 the PR's base predates the current
gateway layout by ~11.9K commits). Messaging platforms can set
gateway.platforms.<key>.skip_context_files: true to skip the
filesystem-heavy context-file discovery (SOUL.md, AGENTS.md,
.cursorrules walks) during AIAgent construction \u2014 10-100x slower
stat()/walk costs on Windows made this a real per-turn tax. Soul
identity is still loaded (single small file), so the persona survives.
The flag participates in _agent_config_signature so toggling it
rebuilds the cached agent instead of silently reusing a prompt built
under the other setting (prompt-cache correctness).
The PR's hunk 1 (mtime-caching the per-turn dotenv reload) was dropped:
df51ad797 mtime-cached load_config/read_raw_config and c2eda92fd
removed the per-turn deepcopies, capturing most of that win; the
function has since gained a multiplex early-return and managed-scope
overlay that the original whole-function skip would have bypassed.
Review folds on the #60807 salvage:
- resolve_skin tests are behavioral (thread-ident probe + ready-frame
wiring check) instead of pure source inspection, per the #72720
pattern; a source assertion remains as belt-and-braces.
- The warm-list test does REAL imports and checks sys.modules —
_warm_gateway_module swallows ImportError by design, so the PR's
tracking-stub test would pass even with a typo'd module name.
- resolve_copilot_token logs a debug line when the env-var
short-circuit skips the gh-CLI fallback (behavioral change made
observable).
Three fixes for the Desktop/TUI cold-start stall where the event loop
is blocked for ~14s between HERMES_BACKEND_READY and the first
prompt (#60800):
1. copilot_auth: skip subprocess fallback when any
Copilot env var is explicitly set (even if invalid). The user
expressed token intent via env var; silently substituting a CLI
token is surprising and the subprocess adds up to 5s on Windows.
2. tui_gateway/ws: run resolve_skin() via asyncio.to_thread so config
loading + skin engine init do not block the WS read loop during
the cold-start RPC burst.
3. web_server: extend _warm_gateway_module to pre-import the heavy
module chains (auth, copilot_auth, runtime_provider, skin_engine,
inventory, model_switch) that the first WS connection + RPC burst
would otherwise import on the loop thread. These trigger .pyc
compilation and Defender scans on Windows (15-30s per the existing
comment) and were not covered by the original gateway-only warm.
Tests: 5 new tests in test_cold_start_gil_stall.py + 2 new tests in
test_copilot_auth.py. All 36 copilot_auth tests + 16 ws/web_server
tests pass.
CI slice 3/7 failures: run_conversation tests pass MagicMock base_urls
through the metadata probe path; re.sub raised TypeError where the old
code let non-strings flow through. Preserve that contract.
fetch_endpoint_model_metadata's generic (non-LM-Studio) /models fetch and
its llama.cpp /v1/props context-length follow-up built request URLs
straight from the unrewritten candidate, unlike every other local-probe
site. Both retained the multi-second dual-stack IPv6 connect penalty
that _localhost_to_ipv4() exists to skip (measured on macOS: localhost
32.9ms vs 127.0.0.1 0.1ms on a dead port; ~2s on Windows). normalized
stays the cache key so caching behavior is unchanged; only the outbound
request target is rewritten.
Re-derived from PR #61528 onto current main (original no longer applied
cleanly).
kshitij