Commit Graph

33348 Commits

Author SHA1 Message Date
Victor Kyriazakos 49ef015ca3 fix: join delegated work before finite chat exits 2026-09-10 05:02:41 -07:00
fangliquanflq fa75692211 docs(agent): clarify detached fork lifecycle boundary 2026-09-10 13:23:00 +02:00
fangliquanflq 77f2ce853f fix(agent): suppress detached fork session-end hooks 2026-09-10 13:23:00 +02:00
fangliquanflq feb03196a2 fix(agent): isolate detached forks from lifecycle hooks 2026-09-10 13:23:00 +02:00
Teknium 6964eebd35 fix(deepseek): pass custom model ids through instead of rewriting them
The deepseek provider allow-listed ids by shape (two canonicals plus a
^deepseek-v<N> regex) and rewrote everything else to deepseek-flash. That
swallowed the vendor's own deepseek-flash the day it shipped (#107206) and
would do the same to any future id without a v<N> marker, while still
letting shape-matching guesses such as deepseek-v4.1-flash through to a 400.

Only the two ids DeepSeek actually retired (deepseek-chat, deepseek-reasoner)
are remapped now; every other id the user typed reaches the wire as typed and
the API's own error names the valid models.
2026-09-10 04:22:38 -07:00
hermes-seaeye[bot] 1603a073e4 fmt(js): npm run fix on merge (#107293)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2026-09-10 11:04:48 +00:00
686f6c61 f44b0fd342 fix(desktop): keep model picker on the selected provider
Treat the Desktop catalog selection as a (provider, model) pair so a
shared model id on OpenRouter cannot rewrite a custom or first-party
pick after refresh or keyboard highlight.
2026-09-10 03:58:52 -07:00
Teknium 69a4c42c1d fix(acp): explicit provider:model prefix is never re-detected (#59089)
session/set_model with "anthropic:claude-sonnet-5" while already on anthropic
parsed to the same provider, so the bare-name fallback detect_provider_for_model
ran and could hand the session to OpenRouter because the bare name exists in
its catalog. Detection now runs only when parse_model_input consumed no prefix.

Diagnosed by AlexFucuson9 in #59294; same fix, expressed as "input unchanged
by parsing" so it needs no provider-name table.
2026-09-10 03:58:37 -07:00
Teknium 696d078959 fix(deepseek): only fold deepseek-* spellings onto deepseek-flash
Names outside the family (qwen3.8-max typed on the DeepSeek provider) now
pass through so the validator rejects them, instead of silently running
Flash under a different label.
2026-09-10 03:58:37 -07:00
Teknium 70ed17e3d8 fix(cli): same-provider /model on a session-only custom endpoint stays on it (#74143)
A bare `custom`/`local` session whose base_url is not the trusted config
`model.base_url` re-resolved credentials from config on a same-provider
switch and fell through to the OpenRouter default: the next turn hit
openrouter.ai with an empty (or the custom) key. Keep the session endpoint
and key when resolution comes back empty or on OpenRouter; a config-backed
custom URL still wins so key/endpoint rotation is not pinned to a stale
session.

Diagnosed by fangliquanflq in #74143 / #71693; this is the minimal form of
that fix at the same-provider credential step.
2026-09-10 03:58:37 -07:00
Teknium 2466684db5 fix(models): never auto-switch to a provider the user has no credentials for
/model <name> on provider A, where the name is only known to provider B
(static catalog or OpenRouter), switched the session to B even when B had
no key: an immediate 401 for most vendors, and for OpenRouter — whose
runtime resolves with an EMPTY key instead of raising — a silent switch
onto a metered aggregator. The dashboard's flat Model field had two more
copies of the same guess ("vendor/model on a native provider" → openrouter).

detect_provider_for_model() now walks its ladder as candidates and skips any
target without credentials (env/.env key, auth-store login, or a usable
credential pool entry). Exceptions: the user NAMED the provider (/model nous)
or there is no current provider yet ("auto") — then the guess is handed back
so the credential step fails loudly instead of silently ignoring input. A
vendor/ prefix naming a provider declared in `providers:` is a selection, not
a guess, and always routes. The dashboard fallbacks apply the same gate.

Tests that pinned "switch to OpenRouter/vendor with no key" now grant the
credential they assumed; two new invariants cover the gate.
2026-09-10 03:58:37 -07:00
Teknium a0749d583a refactor(sessions): trim #106543 to the host-side heal and two invariants
The publish-side taxonomy (_end_stamp_class, _compression_parent_obstacle,
compression_parent_deliberately_ended) and the agent-guard delegation produced
exactly main's verdict -- every non-automatic stamp still fails closed -- so
they only changed an error string. Inline the "explicit close with no
continuation" test into reopen_if_explicitly_closed(), restore main's publish
branch and agent guard untouched, and keep two tests: the field shape rotates
after the host clears the stamp; boundary/compression/automatic stamps and a
session already claimed for teardown are never cleared.
2026-09-10 03:04:51 -07:00
Totoro-qaq 9cd1453f26 fix(sessions): a stale explicit-close stamp no longer makes a session permanently uncompressible
Consolidated from PR #106543 (5 commits, final tree d2c4d908) by @Totoro-qaq.
publish_compression_child() fails closed on any non-automatic end stamp and
end_session() is first-stamp-wins, so a stale tui_close on a session the TUI
still routes turned every rotation into "compute the summary, then discard it"
(#106459). The host that still routes the session clears the stale explicit
close via SessionDB.reopen_if_explicitly_closed() before the turn starts;
publication never heals explicit closes. Review probes by @ehz0ah.
2026-09-10 03:04:51 -07:00
kshitijk4poor 4a76e99f87 docs(cron): drop stale mirror-scope wording left by the home lane
- `_target_matches_origin` docstring still claimed fan-out targets are
  "deliberately NOT mirrored" and mirroring is origin-only; that has been
  false since origin_fallback landed and is doubly so with the home lane.
  Point to `_target_mirror_eligible` for the policy instead.
- cron.md DM-only bullet said "origin DM session"; the config comment in
  the same change already says "target DM". Align.
- Remove two `_expand_routing_tokens` unit assertions bolted into the
  eligibility test; the `" ALL , slack "` parametrize covers expansion
  end-to-end through `_resolve_delivery_targets`.
2026-09-10 15:32:47 +05:30
Victor Kyriazakos 22356075f1 fix(cron): make user-written bare-platform home targets continuable
A managed cron with `deliver: "slack"` (no captured origin) delivers to the
Slack home channel, but the brief was never mirrored into that session and
the in_channel seed never fired: bare-platform targets resolved with no
`_resolved_from` provenance, so `_target_mirror_eligible` treated them like
`all` broadcast expansions.

- `_resolve_delivery_targets` derives `from_broadcast` from the raw token
  and passes it to `_resolve_single_delivery_target`; a user-written bare
  platform token is tagged `_resolved_from: "home"`, `all` expansions stay
  untagged (fan-out is never continuable).
- `_target_mirror_eligible`: `home` eligible under the same flags as
  `origin_fallback` (per-job attach_to_session wins, else cron.mirror_delivery).
- `_MIRROR_PROVENANCE_RANK`: `home` == `origin_fallback` so token order
  through dedup cannot strip eligibility.
- Docs, tool schema text, cron/gateway AGENTS.md updated; existing
  exact-dict pins carry the new key.

Squashed from PR #101819 (commits 5fe612c9, 65e36926, 10e0af1d, f2cc4639):
they predate the cron/scheduler.py -> scheduler_delivery.py split and do
not cherry-pick individually onto current main.
2026-09-10 15:32:47 +05:30
hermes-seaeye[bot] 16a408534c fmt(js): npm run fix on merge (#107255)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2026-09-10 09:55:33 +00:00
Teknium 6271d772d2 docs: Telegram Quick setup on Desktop and the Messaging restart banner 2026-09-10 02:48:21 -07:00
Teknium d5495e762c fix(desktop): messaging restart banner lives in the detail column; drop duplicate toast CTA
Live pass showed the banner spanning list + detail and the save toast carrying a
second 'Restart gateway' button beside the banner's 'Restart now'. Render the
banner inside DetailColumn (content width, above the platform header) and let the
toast stay a plain confirmation. Also hide the redundant 'Messaging gateway
stopped' setup pill when the state pill already says exactly that.
2026-09-10 02:48:21 -07:00
Teknium 52dea2c7d9 feat(desktop): Telegram QR quick setup + persistent restart banner on Messaging
Port the dashboard Channels flow into the Desktop Messaging page:

- Telegram "Quick setup": the Nous pairing service mints a bot; the page shows
  the QR / deep link, polls until Telegram confirms, pre-fills the detected
  owner id into the allowlist, and applies. The backend writes the token +
  allowlist into the scoped profile and restarts the gateway; the page watches
  the restart child's exit so a failed restart surfaces instead of a silent
  "gateway stopped".
- Restart-needed banner: every credential save / clear / enable toggle keeps a
  banner with "Restart now" up until a restart actually completes. Toasts
  vanish; the credential still only loads on the next gateway start.
- runGatewayRestart now resolves a boolean (never rejected before, so the
  Webhooks page's catch path was dead and its banner cleared on failure).
- New API helpers carry the page's profile scope on every pairing call so
  start/status/apply hit the same backend that holds the pairing in memory.
- i18n keys in all six locales; `qrcode` added to the desktop deps (already
  pinned identically in web/).
2026-09-10 02:48:21 -07:00
Teknium 7f487df266 fix(gateway): hermes gateway restart without a service unit no longer bails on linger
The linger hint explains a FAILED systemd unit restart. It ran unconditionally,
so on any Linux login session with linger off and no unit installed the
command printed the hint and exited 0 without stopping or starting anything.
The dashboard/Desktop "Restart gateway" action spawns exactly this command and
polls its exit code, so it reported success while the gateway never came back
with the newly saved credentials. Gate the hint on an installed systemd unit;
the no-unit path falls through to the detached restart as intended.
2026-09-10 02:48:21 -07:00
Teknium e1f1c8a935 fix(dashboard): stopped gateway no longer reports a dead process's platform error
gateway_state.json preserves per-platform entries across restarts, so a gateway
that once ran without TELEGRAM_BOT_TOKEN and then stopped kept reporting
'fatal / No bot token configured' on the Channels/Messaging pages even after the
user saved a token — the Desktop showed 'Saved' on both fields next to the error.
Only a live gateway's verdict describes current config; when no gateway is running
the platform reads gateway_stopped with no error.
2026-09-10 02:48:21 -07:00
Teknium 2bbb279c18 chore: map contributor email for salvaged #107126 (MiseHinoha) 2026-09-10 02:44:26 -07:00
Teknium aeecb110f8 fix(deepseek): deepseek-flash is the canonical Flash id; retired names fold onto it
DeepSeek retired deepseek-v4-flash on 2026-09-10 (V4.1-Flash release); the API's
model name is now `deepseek-flash` and /v1/models lists only it. Hermes still
folded every non-V-series name onto deepseek-v4-flash, so `/model deepseek-flash`
on the DeepSeek provider was rewritten, then the validator "auto-corrected" it
back against the live listing: "Auto-corrected deepseek-v4-flash -> deepseek-flash"
on every switch.

Retired aliases (deepseek-chat / -reasoner and other fuzzy names) now fold onto
deepseek-flash; the curated catalog, profile fallback list, aux default, goal-judge
hint and pricing snapshot (2026-09-10 off-peak USD) follow the docs. Dated
deepseek-v4-* ids still pass through untouched.

Builds on YipTszkwan's #107126 (earliest fix in the cluster).
2026-09-10 02:44:26 -07:00
Teknium 215fd0ecb9 fix(models): a model the current provider serves never re-routes to OpenRouter
detect_provider_for_model() walked static catalogs and then the OpenRouter
catalog. Providers whose live catalog outruns the static list (Codex
early-access ids, Nous Portal slugs, Ollama Cloud) had nothing to stop the
ladder, so `/model gpt-6-astra` from a Codex session, or `/model glm-5.3-flash`
from a Nous-sub-only session, silently rebuilt the agent on metered OpenRouter
(or on a vendor the user has no key for).

The current provider's disk-cached live catalog (cached_provider_model_ids, 1h
TTL + SWR) is now consulted first: an exact or bare-name hit stays put and
resolves to the provider's own spelling. Fixes the whole class at the shared
choke point, covering /model, ACP, oneshot and the dashboard.

Refs #97487 (sj-unit72 diagnosed the ollama-cloud instance of this).
2026-09-10 02:44:26 -07:00
YipTszkwan 8435a3ae00 fix(deepseek): recognise the version-less deepseek-flash model id
DeepSeek's 2026-09 Flash refresh introduced a version-less canonical id:
GET /v1/models now returns `deepseek-flash` (alongside `deepseek-v4-pro`), the
API accepts it directly, and the older `deepseek-v4-flash` is server-side
aliased onto it. Every DeepSeek model-id gate in Hermes keys off the
`deepseek-v<N>` prefix, so the new id silently missed all four:

* DeepSeekProfile.build_api_kwargs_extras classified it as non-thinking and
  omitted `extra_body.thinking`. The server then defaults to thinking-on, so
  the user's thinking toggle and `reasoning_effort` were quietly ignored.
* `_normalize_for_deepseek` folded it onto `deepseek-v4-flash` (it misses the
  V-series regex), so the id a user picked never reached the wire and the
  config stored a different model than the picker advertised.
* `DEFAULT_CONTEXT_LENGTHS` fell through to the 128K `deepseek` catch-all
  instead of the real 1M window, capping the model at an eighth of its
  context before compaction kicked in.
* `_REASONING_STALE_TIMEOUT_FLOORS` had no entry, leaving the stale-stream
  detector at its 180s default instead of the 600s reasoning-model floor.

Verified live against api.deepseek.com: `deepseek-flash` answers 200 with
`model: deepseek-flash`, accepts image input (the refresh folds vision into
the Flash model), and the in-between id `deepseek-v4.1-flash` is rejected
with "The supported API model names are deepseek-flash, deepseek-v4-pro".

Adds the id to all four gates plus regression coverage for each site.
2026-09-10 02:44:26 -07:00
hermes-seaeye[bot] 5172f22df2 fmt(js): npm run fix on merge (#107244)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2026-09-10 09:37:09 +00:00
Teknium 61afcde8f9 refactor(desktop): one Plugins surface — Capabilities → Plugins owns agent + desktop plugins, install, and the catalog
Plugins were split across two pages that each showed half the picture:
Settings → Plugins listed desktop plugins plus "Install from Git" and a
pointer saying agent plugins live elsewhere; Capabilities → Plugins listed
agent plugins plus the catalog picker but knew nothing about desktop
plugins. A user asking "what extends my Hermes and where do I add more?"
had to visit both and still could not see the whole set in one place.

Capabilities → Plugins is now THE plugins page:

- Agent plugins section (scoped to the profile selector) with the
  "Install from Git" button in its header — installs target the scoped
  profile, not whichever one is active.
- Desktop plugins section beneath it (same for every profile), with the
  folder/rescan controls and the "agent half missing here" drift chip,
  whose repair also lands in the SCOPED profile.
- The catalog picker underneath, unchanged.

Settings → Plugins is removed. `/settings?tab=plugins[&plugin=…]` and the
existing `?tab=mcp` redirect share one table (`settings/moved-tabs.ts`) so
old bookmarks and palette links land on the same row on the new page.
Command palette: plugins moved from the Settings group to the Capabilities
group; installed-plugin rows deep-link to `/skills?tab=plugins&plugin=…`.
Dead `settings.plugins.agent.*` and `settings.nav.plugins` i18n keys dropped;
docs and in-code pointers say Capabilities → Plugins.
2026-09-10 02:30:50 -07:00
Teknium 8068c09432 fix(desktop): verify the Windows update receipt against the checkout, not cwd
The Windows hand-off script runs from the PRE-update checkout, spawned from
HERMES_HOME by the Desktop. e7eaff68 shipped the post-update verify step
without the cwd pin that landed a day later, so verify_windows_desktop_update(Path.cwd())
looked for apps/desktop/release under ~/.hermes and reported a healthy,
fully updated install as "The updated Desktop executable is missing" (exit 8,
error dialog, app relaunched fine).

Derive the root from the imported hermes_cli package instead; the ps1 no
longer passes a cwd. The cwd pin stays as belt for the other child steps.
2026-09-10 01:25:45 -07:00
Teknium 4baf4269ff fix(metadata): GLM context windows from the live catalog; hyphenated slugs, entry-level overrides, aux inherits main pin
Every GLM slug that missed a DEFAULT_CONTEXT_LENGTHS key fell to the "glm" 202,752 catch-all, so
compression fired at ~20% of the real window and the aux feasibility check auto-lowered the session
threshold to that number (the 202,752 in the Coatue compression report).

- Catalog: GLM entries from Nous + OpenRouter /v1/models (2026-09-09): 5.3 / 5.3-flash 1,310,720
  (:batch/:US 1,048,576), 5.2 1,048,576, 5 / 5.1 / 4.7 / 4.6 204,800, *-turbo / 4.7-flash 202,752.
- _catalog_key_matches: version separators normalised on both sides so relay slugs like z-ai-glm-5-3
  hit glm-5.3 (#97398; approach from PR #97412 by @shellybotmoyer, re-implemented on current main).
- get_custom_provider_context_length: entry-level custom_providers[].context_length backs every model the
  entry serves when no per-model override exists, so /model switch stops dropping it (#98387; from
  PR #98396 by @liuhao1024, re-implemented on the config_providers sibling).
- check_compression_model_feasibility: when aux compression is the main model on the main route, reuse the
  main model's resolved window instead of re-resolving without its pin (#89500 mechanism 2, #45519).

Closes #97398, #98387, #89500, #87825, #97595, #97820 (suffix strip already on main; catalog values now match).
2026-09-10 01:08:18 -07:00
Teknium 2554a16a1d fix(desktop/settings): let searchable-select popovers size to content, not the shrink-wrapped trigger
The Timezone field (and the voice/model ComboboxInput) pinned the popover to
--radix-popover-trigger-width. In the settings grid the trigger shrink-wraps to
its current value (~90-120px), so every IANA row was clipped to "Africa/A…" and
the picker was unusable by eye even though selection itself worked. Use min-w so
the list keeps its own width and only grows to cover a wider trigger.

Live-verified on a sandboxed Electron instance over CDP: popover 123px -> 288px,
rows no longer overflow; click, type+Enter and autosave to config.yaml all land.
2026-09-10 01:04:59 -07:00
Teknium d783c312a7 fix(plugins): run gh auth token under the noninteractive git env
The MCP-catalog noninteractive contract test asserts every subprocess spawned during a git
install carries GIT_TERMINAL_PROMPT=0 and a closed stdin; the gh token probe was spawned with
the inherited env. Use the same hardened env (plus GH_PROMPT_DISABLED) so gh cannot open a
browser/device flow either, and let the contract accept a stdin fed by input= (credential fill
writes its request and closes).
2026-09-10 01:00:19 -07:00
Teknium 9886f6e53b feat(plugins): install plugins from private git repos using the user's stored credentials
`hermes plugins install owner/private-repo` failed for every private repository
even when the same user could `git clone` it from their shell: the hardened
`noninteractive_git_env` disables credential helpers, askpass and global config
(so a hostile repo can't make our plumbing prompt or hang), which also blocks
the user's own stored credential. The result was "could not read Username" or a
60s hang on a GUI askpass, with no hint about how to authenticate.

New `hermes_cli/git_credentials.py` resolves a credential up front from sources
the user already owns — GITHUB_TOKEN/GH_TOKEN (profile-scoped), `gh auth token`,
then `git credential fill` against their configured helpers with prompting
disabled (any host) — and hands it to git as a one-shot
`http.<origin>/.extraheader` via the GIT_CONFIG_* env block. Nothing lands in
the URL, `.git/config` or install metadata. The same path covers
`plugins update`, catalog MCP git installs and profile-distribution staging,
which share the same hardened env and the same failure.

A private-repo clone with no credential now fails fast with an actionable hint.
2026-09-10 01:00:19 -07:00
brooklyn! 6e07eb4838 docs(desktop): define panel titlebar and sticky backdrop behavior 2026-09-10 01:14:30 -05:00
brooklyn! e7c819a7e1 fix(desktop): mask text behind sticky user messages 2026-09-10 01:14:30 -05:00
brooklyn! bbe212de9b feat(desktop): keep panel tabs in the titlebar 2026-09-10 01:14:30 -05:00
hermes-seaeye[bot] cfdbbb6e35 fmt(js): npm run fix on merge (#107102)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2026-09-10 04:49:08 +00:00
hermes-seaeye[bot] 425b00174d fmt(js): npm run fix on merge (#107101)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2026-09-10 04:43:55 +00:00
hermes-seaeye[bot] 7cdc6a7338 fmt(js): npm run fix on merge (#107097)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2026-09-10 04:38:38 +00:00
brooklyn! 78052fe437 fix(desktop): keep notification IPC types independent of the renderer
The Electron composite project imported src/global.d.ts and followed its renderer-only dependencies. Share the notification payload from a dependency-free Electron module instead; all desktop typechecks and lint now pass.
2026-09-09 23:37:48 -05:00
brooklyn! 85571c16b6 fix(desktop): reveal the notification session without replacing its tab
Resolve missing runtime mappings through the shared mirror, dismiss covering routes, and preserve Bot tab ownership. Includes the mapping fix from https://github.com/NousResearch/hermes-agent/pull/100405 (b9e66c4a485dc353238fab8c7a07aa16107d11b7).

Co-authored-by: Nikita Simakov <everm1nd.mail@gmail.com>
2026-09-09 23:37:48 -05:00
brooklyn! 7107458d96 fix(desktop): route native notifications back to their source window
Capture durable focus identity before runtime bindings disappear, retain native handlers, and keep runtime approval actions and plugin callbacks on their source renderer.
2026-09-09 23:37:48 -05:00
Brian Sweatt f557829bff fix(desktop): retain notification handlers across banner dismissal
Adapted from https://github.com/NousResearch/hermes-agent/pull/65897 (a6fb6fe063e235c33b41212bd9cdaea4f9dcb1dc). Keep handlers across close, release on consumption or failure, and dismiss expired notifications before releasing them.

Co-authored-by: brooklyn! <brooklyn.bb.nicholson@gmail.com>
2026-09-09 23:37:48 -05:00
outdog-hwh a245d9aaa4 fix(desktop): resolve native notification action indexes
Adapted from https://github.com/NousResearch/hermes-agent/pull/51893 (a70b0991b2d72796479d63532ccf0fac15b17ba0).

Co-authored-by: brooklyn! <brooklyn.bb.nicholson@gmail.com>
2026-09-09 23:37:48 -05:00
brooklyn! c7676fc6b7 docs(desktop): explain radio artist lookups and privacy 2026-09-09 23:32:55 -05:00
brooklyn! 983f0c2870 feat(desktop): link radio credits directly to artist profiles 2026-09-09 23:32:55 -05:00
brooklyn! a6102b8d80 docs(desktop): document Radio controls and privacy 2026-09-09 22:57:27 -05:00
brooklyn! 3f53719c1d feat(desktop): ship Radio as an opt-in SDK plugin 2026-09-09 22:57:27 -05:00
hermes-seaeye[bot] 1675f1f2c2 fmt(js): npm run fix on merge (#107057)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2026-09-10 03:02:40 +00:00
brooklyn! b5c0a7ebe2 test(desktop): cover clarify submit shortcuts and input guards 2026-09-09 21:57:01 -05:00
brooklyn! ff7a1e8953 fix(desktop): restore Cmd/Ctrl+Enter in clarify forms 2026-09-09 21:57:01 -05:00