Commit graph

17262 commits

Author SHA1 Message Date
Alex Fournier
e9b5e4c8f4 fix(relay): protect turn initialization cleanup
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:48:52 -07:00
Alex Fournier
937fffcfec fix(smoke): resolve Hermes across environments
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:45:40 -07:00
Alex Fournier
1500ce163c fix(relay): normalize rewritten LLM responses
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:44:27 -07:00
Alex Fournier
a54e52aeb1 fix(tools): preserve sequential middleware trace
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:43:27 -07:00
Alex Fournier
0be07b0553 fix(middleware): hide Relay control flag
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:41:47 -07:00
Alex Fournier
2378bd4e4c fix(tools): prevent duplicate managed dispatch
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:40:44 -07:00
Alex Fournier
ea171c947a perf(relay): bypass inactive tool interception
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:38:49 -07:00
Alex Fournier
36185bf2e2 feat(telemetry): expose shared metrics setup
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:36:00 -07:00
Alex Fournier
591ba267d2 fix(relay): release session lock before callbacks
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:33:41 -07:00
Alex Fournier
e862099dff fix(relay): merge Anthropic stream usage
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:32:44 -07:00
Alex Fournier
c789bdd38a fix(relay): close managed provider streams
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:32:03 -07:00
Alex Fournier
f370af6816 fix(codex): stop draining interrupted streams
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:26:49 -07:00
Alex Fournier
9dfe690125 fix(relay): preserve provider request extensions
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:24:19 -07:00
Alex Fournier
a6fbff1d7e fix(relay): retain active child sessions
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:22:50 -07:00
Alex Fournier
398a8ca580 fix(tools): serialize concurrent approvals
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:19:03 -07:00
Alex Fournier
c16f86dc4c fix(relay): serialize rewritten tool results
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:13:58 -07:00
Alex Fournier
1c6a96032c fix(relay): preserve successful managed results
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:12:50 -07:00
Alex Fournier
1ffeaf226c test(relay): skip native suites without binding
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 13:04:12 -07:00
Alex Fournier
c9747106b6 fix(runtime): retry Relay stream construction failures
Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-23 07:22:38 -07:00
Alex Fournier
b4e105031a Merge upstream main into feat/hermes-relay-shared-metrics
# Conflicts:
#	MANIFEST.in
#	pyproject.toml
#	tests/test_project_metadata.py
2026-07-23 07:22:03 -07:00
wz-heng
91546b8337 fix: preserve named custom provider vision overrides 2026-07-23 17:57:33 +05:30
kshitij
97d51ca20d fix: use canonical _get_platform_tools resolver for memory tool status
The PR's inline toolset resolution (checking 'memory' in cli_toolsets
list) produced wrong results for composite toolsets like 'hermes-cli'
which expand to include the memory tool. Replace with the canonical
_get_platform_tools() from tools_config.py which correctly handles
composite toolsets and all edge cases.

Update tests to mock _get_platform_tools instead of raw config.
2026-07-23 17:50:20 +05:30
kshitij
c82c5786f4 test: cover config-aware memory status labels + AUTHOR_MAP entry
Add tests/hermes_cli/test_memory_status.py with 11 tests covering:
- No hardcoded 'always active' label
- memory_enabled, user_profile_enabled, memory toolset indicators
- Tool enabled/disabled via platform_toolsets.cli
- Provider still shown alongside indicators

Add huajiang@tubi.tv → thirstycrow to AUTHOR_MAP (PR #23630 salvage).
2026-07-23 17:50:20 +05:30
Hua Jiang
4c99a44e6d fix: hermes memory status now reads actual config instead of hardcoded 'always active'
The 'Built-in: always active' label was a hardcoded string that never
reflected the user's actual configuration. It now shows three separate
indicators, each reading from the real source of truth:

  - Memory injection:  reads memory.memory_enabled from config.yaml
  - User profile:      reads memory.user_profile_enabled from config.yaml
  - Memory tool:       checks if 'memory' is in platform_toolsets.cli
                       (or defaults to enabled if no explicit list)

Before:
  Built-in:  always active

After:
  Built-in (MEMORY.md / USER.md):
    Memory injection:   disabled ✗
    User profile:       disabled ✗
    Memory tool:        disabled ✗
2026-07-23 17:50:20 +05:30
kshitijk4poor
2841a9cbca chore: add contributor email mapping for fangliquanflq 2026-07-23 17:38:05 +05:30
Fangliquan
9220c0c0bb fix(agent): close tool-result tails on invalid-tool and truncated-tool early returns
Invalid-tool exhaustion and truncated-tool early returns skipped finalize_turn, leaving role=tool transcripts that become tool→user on the next turn for strict providers. Call close_interrupted_tool_sequence before persist on those paths (same as interrupt aborts).
2026-07-23 17:38:05 +05:30
kshitij
8058e01834 refactor: dedupe check_info call + fix trailing whitespace in doctor
Hoist the duplicated check_info(source_id) call out of both
if/else branches into a single call after the branch. Remove
trailing whitespace on the blank line after the except block.

Follow-up cleanup for PR #69981.
2026-07-23 17:32:38 +05:30
kshitij
7419de6ac6 refactor: simplify WAL-reset gate warning dedup + tuple handling
Consolidate the two near-identical warning strings in
_log_wal_reset_bug_once into a single logger.warning call with an
action variable. Remove overengineered defensive tuple-length handling
in is_sqlite_wal_reset_vulnerable (sqlite3.sqlite_version_info always
returns a 3-tuple). Remove extra blank line.

Follow-up cleanup for PR #69981.
2026-07-23 17:32:38 +05:30
HexLab98
a94f8e69e1 test(state): cover SQLite WAL-reset version gate and doctor probe
Assert the version matrix, fresh-DB DELETE fallback, already-WAL left
alone (no checkpoint/DELETE), fixed-SQLite WAL path, and warn-only
doctor output for vulnerable builds (#69784).
2026-07-23 17:32:38 +05:30
HexLab98
953cbc0300 fix(state): refuse WAL on SQLite builds with the WAL-reset bug
On vulnerable SQLite (e.g. 3.50.4), do not enable WAL for fresh/non-WAL
shared databases — prefer DELETE instead. Leave existing on-disk WAL
alone (no live downgrade under concurrent gateway/cron openers). Surface
Python/SQLite version details as a doctor warning (#69784).
2026-07-23 17:32:38 +05:30
brooklyn!
8fc278207b
Merge pull request #69938 from NousResearch/bb/desktop-dev-fixes
fix(desktop): keep gateway session alive across Vite HMR
2026-07-23 02:22:51 -05:00
hermes-seaeye[bot]
44b171fe3d
fmt(js): npm run fix on merge (#69941)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2026-07-23 07:19:06 +00:00
Brooklyn Nicholson
ecc6aec4bf fix(desktop): isolate gateway HMR survivor across vitest cases
Vitest keeps import.meta.hot truthy, so boot-effect cleanup parks the
open socket; drain it between cases so the next test boots fresh.
2026-07-23 02:15:39 -05:00
Brooklyn Nicholson
cbea2acbb8 fix(desktop): gate HMR session-survival entirely out of prod builds
Guard the globalThis gateway-state container and the survivor park/adopt
calls on the import.meta.hot literal (not the runtime hmrActive() helper),
so Vite dead-code-eliminates every HMR path in production. Prod now uses a
plain module-local singleton — no globalThis, no Symbol.for — and the
survivor module drops out of the bundle entirely. Verified: gatewayRegistryState,
gatewaySurvivor, and import.meta.hot are all absent from the prod build.

Removes the now-unused hmrActive() export.
2026-07-23 02:15:25 -05:00
Brooklyn Nicholson
1943c1b750 fix(desktop): dedup profile in HMR adoptBoot 2026-07-23 02:15:25 -05:00
Brooklyn Nicholson
813ddb4555 fix(desktop): keep gateway session alive across Vite HMR
Park the live primary gateway socket on Fast Refresh dispose and re-adopt it
on remount so dev UI edits don't tear down the WebSocket. Hold gateway store
singletons on globalThis + self-accept HMR on store/gateway.ts. Prod strips
import.meta.hot — live unmount unchanged.
2026-07-23 02:15:25 -05:00
hermes-seaeye[bot]
7c9d05267c
fmt(js): npm run fix on merge (#69939)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
2026-07-23 07:08:53 +00:00
brooklyn!
3e163d29bd
Merge pull request #69936 from NousResearch/bb/voice-speak-whole-turn
fix(voice): speak the whole desktop turn and idle-flush held narration
2026-07-23 02:00:30 -05:00
Brooklyn Nicholson
9859e1f7df fix(voice): speak the whole turn, not just its first bubble; idle-flush held narration
Two fixes for desktop hands-free voice:

- The live speech session bound to the first assistant bubble with text, so
  a tool-calling turn spoke only the opening narration and silently dropped
  every later interim AND the final answer. The conversation selector now
  aggregates all unspoken assistant bubbles in order (turn-scoped speech);
  auto-speak keeps its latest-reply-only behavior.

- The speak-stream WS producer blocked forever on the text queue, so a
  narration line with no trailing whitespace ("Let me check.") sat in the
  sentence chunker until end-of-turn — spoken long after the tool finished,
  with the UI stuck on "Preparing audio…". Mirror the CLI speaker's idle
  flush: sentence-terminated buffers flush after 0.5s of producer silence,
  anything else after ~2s; open <think> blocks are never flushed.
2026-07-23 01:53:37 -05:00
brooklyn!
7168845dc3
Merge pull request #69909 from NousResearch/bb/computer-use-perf
perf(computer_use): cap capture size and cache vision routing
2026-07-23 01:49:05 -05:00
Brooklyn Nicholson
cc1765dce2 Merge remote-tracking branch 'origin/main' into bb/computer-use-perf
# Conflicts:
#	hermes_cli/config.py
#	tools/computer_use/cua_backend.py
2026-07-23 01:39:21 -05:00
brooklyn!
93f8da55cb
Merge pull request #69903 from NousResearch/bb/salvage-53841-no-overlay
fix(computer_use): disable cua-driver overlay by default on macOS/WSL (supersedes #53841)
2026-07-23 01:36:49 -05:00
brooklyn!
d3d989c4c2
feat(desktop): show credit-usage notices as toasts (#69808) (#69828)
* fix(desktop): render agent credit notices as toasts (#69808)

The desktop renderer had no handler for the `notification.show` /
`notification.clear` WS events, so every credit-usage notice the backend
sends (`agent/credits_tracker.py` → `tui_gateway/server.py`) was silently
dropped. Credit warnings like "• Credits 50% used · $220.00 cap" never
appeared, even though the Ink TUI renders them in its status bar.

Add the two missing branches to the gateway-event dispatcher, delegating
to a small, pure-testable module:

- `store/agent-notices.ts` — `noticeToToast()` maps a notice to a toast
  (level → toast kind, sticky → durationMs 0, ttl → ttl_ms), and uses the
  notice `key` as the toast id. Re-emitting the same key REPLACES the
  toast, so the credits 50→75→90 line escalates in place instead of
  stacking, and a key-matched `notification.clear` maps straight to
  `dismissNotification(key)`.
- The notice `text` already carries its own glyph (• ⚠ ✕ ✓), so no toast
  icon is added.
- Notices are account-wide, so the toast shows regardless of which
  session is focused.

The Ink TUI (`ui-tui/src/app/turnController.ts`) is the reference for the
latest-wins / sticky-vs-ttl / key-matched-clear behavior.

Export `NotificationInput` so the mapping's return type can be named.

* feat(desktop): native OS credit alerts + billing-page nudge (#69808)

Round out the credit-notice handling from the previous commit with the two
optional pieces from the issue:

- Native OS notification for the urgent pair. `credits.depleted` /
  `credits.restored` also fire an Electron notification when Hermes is
  backgrounded, via a new `credits` NativeNotificationKind (the existing
  five didn't fit) with its own toggle in Settings → Notifications (the
  panel is data-driven off NATIVE_NOTIFICATION_KINDS, so the toggle and
  i18n are the only additions). The escalating usage line and grant-spent
  notice stay in-app toasts only. Dispatch is `global` (account-wide, not
  session-bound) and gated by the user's prefs + backgrounded check.
- Billing-page nudge. A `credits.*` crossing invalidates the
  `['billing','state']` query so Settings → Billing reflects the change
  immediately instead of waiting up to 30s for its poll.

`nativeNoticeInput()` is a pure mapping (urgent-key gate → native input),
unit-tested directly; the gateway-event branch does the localized-title
lookup and gated dispatch. i18n added for all four locales.

* feat(credits): report $used of $cap instead of % in the usage notice

The usage gauge is Nous subscription-cap-only (used_fraction requires a cap;
non-Nous providers emit no headers, so no notice fires). A bare percentage
implied a universal unit that doesn't exist, so report the absolute dollars
used of the cap instead: used = cap - remaining, from micros (money-safe),
clamped to [0, cap]. Still a snapshot at band-crossing (re-emits on band
change, not every turn) to keep the single escalating line and stay quiet on
append-only surfaces (messaging pushes one message per crossing).

* fix(desktop): de-dupe credit toast icon, band-color the figure, split detail

Three fixes to how agent credit notices render as toasts:
- Strip the leading severity glyph (the toast already draws a kind icon, so the
  raw text doubled it). Native OS notifications keep the glyph (no icon there).
- Icon top-margin is now 0.42ch (font-relative) instead of a fixed rem.
- Band-color the $used figure (semibold) by $used/$cap: muted <75%,
  --ui-orange >=75%, --ui-red >=90% (depleted red, restored green), reusing the
  existing --ui-* usage palette. Icon shares the accent.
- Split a trailing '. detail' into a muted secondary line (title+description
  convention) instead of an inline middot.
Generic 'accentColor' + 'meta' slots on the notification; degrades gracefully
when a notice has no figure.

* fix(desktop): billing page always fetches fresh state (team-account desync)

The x-nous-credits-* headers are best-effort and can drift out of sync,
notably in team/org accounts where another member's spend moves the shared
balance without touching this client's headers. The billing endpoint is the
source of truth, so the page no longer trusts a cache: staleTime 0 +
refetchOnMount 'always' force a fresh fetch on every open and focus (still
polling 30s while mounted). The credits.* invalidation nudge still pulls a
crossing in immediately.

* chore(desktop): dev-only credit-notice demo hotkey

Ctrl+Shift+C (and window.__creditsDemo()) steps the full credit-notice
lifecycle (usage 50->75->90, grant-spent, depleted/restored) through the real
gateway event fan-out via a new emitLocalGatewayEvent, so the toast/native/
billing-invalidation paths are testable without hitting real usage bands.
Installed only under import.meta.env.DEV, so it's tree-shaken from production.
2026-07-23 01:36:28 -05:00
Brooklyn Nicholson
8cfc9e4e27 chore(desktop): dev-only credit-notice demo hotkey
Ctrl+Shift+C (and window.__creditsDemo()) steps the full credit-notice
lifecycle (usage 50->75->90, grant-spent, depleted/restored) through the real
gateway event fan-out via a new emitLocalGatewayEvent, so the toast/native/
billing-invalidation paths are testable without hitting real usage bands.
Installed only under import.meta.env.DEV, so it's tree-shaken from production.
2026-07-23 01:28:59 -05:00
Brooklyn Nicholson
664b131c30 fix(desktop): billing page always fetches fresh state (team-account desync)
The x-nous-credits-* headers are best-effort and can drift out of sync,
notably in team/org accounts where another member's spend moves the shared
balance without touching this client's headers. The billing endpoint is the
source of truth, so the page no longer trusts a cache: staleTime 0 +
refetchOnMount 'always' force a fresh fetch on every open and focus (still
polling 30s while mounted). The credits.* invalidation nudge still pulls a
crossing in immediately.
2026-07-23 01:28:59 -05:00
Brooklyn Nicholson
d010220588 fix(desktop): de-dupe credit toast icon, band-color the figure, split detail
Three fixes to how agent credit notices render as toasts:
- Strip the leading severity glyph (the toast already draws a kind icon, so the
  raw text doubled it). Native OS notifications keep the glyph (no icon there).
- Icon top-margin is now 0.42ch (font-relative) instead of a fixed rem.
- Band-color the $used figure (semibold) by $used/$cap: muted <75%,
  --ui-orange >=75%, --ui-red >=90% (depleted red, restored green), reusing the
  existing --ui-* usage palette. Icon shares the accent.
- Split a trailing '. detail' into a muted secondary line (title+description
  convention) instead of an inline middot.
Generic 'accentColor' + 'meta' slots on the notification; degrades gracefully
when a notice has no figure.
2026-07-23 01:28:59 -05:00
Brooklyn Nicholson
59a735b8f3 feat(credits): report $used of $cap instead of % in the usage notice
The usage gauge is Nous subscription-cap-only (used_fraction requires a cap;
non-Nous providers emit no headers, so no notice fires). A bare percentage
implied a universal unit that doesn't exist, so report the absolute dollars
used of the cap instead: used = cap - remaining, from micros (money-safe),
clamped to [0, cap]. Still a snapshot at band-crossing (re-emits on band
change, not every turn) to keep the single escalating line and stay quiet on
append-only surfaces (messaging pushes one message per crossing).
2026-07-23 01:28:58 -05:00
Brooklyn Nicholson
69b97a97f7 Merge remote-tracking branch 'origin/main' into bb/salvage-53841-no-overlay
# Conflicts:
#	tools/computer_use/cua_backend.py
2026-07-23 01:26:21 -05:00
Brooklyn Nicholson
b8a2b9b93e feat(desktop): native OS credit alerts + billing-page nudge (#69808)
Round out the credit-notice handling from the previous commit with the two
optional pieces from the issue:

- Native OS notification for the urgent pair. `credits.depleted` /
  `credits.restored` also fire an Electron notification when Hermes is
  backgrounded, via a new `credits` NativeNotificationKind (the existing
  five didn't fit) with its own toggle in Settings → Notifications (the
  panel is data-driven off NATIVE_NOTIFICATION_KINDS, so the toggle and
  i18n are the only additions). The escalating usage line and grant-spent
  notice stay in-app toasts only. Dispatch is `global` (account-wide, not
  session-bound) and gated by the user's prefs + backgrounded check.
- Billing-page nudge. A `credits.*` crossing invalidates the
  `['billing','state']` query so Settings → Billing reflects the change
  immediately instead of waiting up to 30s for its poll.

`nativeNoticeInput()` is a pure mapping (urgent-key gate → native input),
unit-tested directly; the gateway-event branch does the localized-title
lookup and gated dispatch. i18n added for all four locales.
2026-07-23 01:25:58 -05:00
Brooklyn Nicholson
58e3d41582 fix(desktop): render agent credit notices as toasts (#69808)
The desktop renderer had no handler for the `notification.show` /
`notification.clear` WS events, so every credit-usage notice the backend
sends (`agent/credits_tracker.py` → `tui_gateway/server.py`) was silently
dropped. Credit warnings like "• Credits 50% used · $220.00 cap" never
appeared, even though the Ink TUI renders them in its status bar.

Add the two missing branches to the gateway-event dispatcher, delegating
to a small, pure-testable module:

- `store/agent-notices.ts` — `noticeToToast()` maps a notice to a toast
  (level → toast kind, sticky → durationMs 0, ttl → ttl_ms), and uses the
  notice `key` as the toast id. Re-emitting the same key REPLACES the
  toast, so the credits 50→75→90 line escalates in place instead of
  stacking, and a key-matched `notification.clear` maps straight to
  `dismissNotification(key)`.
- The notice `text` already carries its own glyph (• ⚠ ✕ ✓), so no toast
  icon is added.
- Notices are account-wide, so the toast shows regardless of which
  session is focused.

The Ink TUI (`ui-tui/src/app/turnController.ts`) is the reference for the
latest-wins / sticky-vs-ttl / key-matched-clear behavior.

Export `NotificationInput` so the mapping's return type can be named.
2026-07-23 01:25:58 -05:00