fix: browser_vision ignores auxiliary.vision.timeout config by teknium1 · Pull Request #2901 · NousResearch/hermes-agent

teknium1 · 2026-03-25T02:02:56Z

Summary

browser_vision called call_llm() without passing a timeout parameter, so it always used the hardcoded 30-second default in auxiliary_client.py. This made vision analysis with local models (llama.cpp, ollama) impossible — the OpenAI SDK's httpx client would abort the connection after exactly 30 seconds, regardless of what the user set in config.yaml.

What changed

browser_tool.py: browser_vision() now reads auxiliary.vision.timeout from config (same key vision_analyze already uses) and passes it to call_llm(timeout=...)
Both tools: Default vision timeout bumped from 30s to 120s. 30s is too aggressive for local models running on consumer GPUs.

The timeout chain

The user reported setting command_timeout: 250 and auxiliary.vision.timeout: 250 but llama.cpp still got cancelled after exactly 30 seconds. The issue:

command_timeout only controls the browser screenshot capture, not the LLM call
auxiliary.vision.timeout was read by vision_analyze but NOT by browser_vision
call_llm() defaults to timeout=30.0 → httpx aborts → llama.cpp sees "cancel task"

Config

Users can set auxiliary.vision.timeout in config.yaml:

auxiliary:
  vision:
    timeout: 250  # seconds

Test plan

2300 tests pass, 0 failures
Config path verified consistent with vision_tools.py implementation

Fixes user report from GamerGB1988.

… session:end event The hooks page only documented gateway event hooks (HOOK.yaml system). The plugins page listed plugin hooks (pre_tool_call, etc.) that weren't referenced from the hooks page, which was confusing. Changes: - hooks.md: Add overview table showing both hook systems - hooks.md: Add Plugin Hooks section with available hooks, callback signatures, and example - hooks.md: Add missing session:end gateway event (emitted but undocumented) - hooks.md: Mark pre_llm_call, post_llm_call, on_session_start, on_session_end as planned (defined in VALID_HOOKS but not yet invoked) - hooks.md: Update info box to cross-reference plugin hooks - hooks.md: Fix heading hierarchy (gateway content as subsections) - plugins.md: Add cross-reference to hooks page for full details - plugins.md: Mark planned hooks as (planned)

browser_vision called call_llm() without passing a timeout parameter, so it always used the 30-second default in auxiliary_client.py. This made vision analysis with local models (llama.cpp, ollama) impossible since they typically need more than 30s for screenshot analysis. Now browser_vision reads auxiliary.vision.timeout from config.yaml (same config key that vision_analyze already uses) and passes it through to call_llm(). Also bumped the default vision timeout from 30s to 120s in both browser_vision and vision_analyze — 30s is too aggressive for local models and the previous default silently failed for anyone running vision locally. Fixes user report from GamerGB1988.

* docs: unify hooks documentation — add plugin hooks to hooks page, add session:end event The hooks page only documented gateway event hooks (HOOK.yaml system). The plugins page listed plugin hooks (pre_tool_call, etc.) that weren't referenced from the hooks page, which was confusing. Changes: - hooks.md: Add overview table showing both hook systems - hooks.md: Add Plugin Hooks section with available hooks, callback signatures, and example - hooks.md: Add missing session:end gateway event (emitted but undocumented) - hooks.md: Mark pre_llm_call, post_llm_call, on_session_start, on_session_end as planned (defined in VALID_HOOKS but not yet invoked) - hooks.md: Update info box to cross-reference plugin hooks - hooks.md: Fix heading hierarchy (gateway content as subsections) - plugins.md: Add cross-reference to hooks page for full details - plugins.md: Mark planned hooks as (planned) * fix: browser_vision ignores auxiliary.vision.timeout config browser_vision called call_llm() without passing a timeout parameter, so it always used the 30-second default in auxiliary_client.py. This made vision analysis with local models (llama.cpp, ollama) impossible since they typically need more than 30s for screenshot analysis. Now browser_vision reads auxiliary.vision.timeout from config.yaml (same config key that vision_analyze already uses) and passes it through to call_llm(). Also bumped the default vision timeout from 30s to 120s in both browser_vision and vision_analyze — 30s is too aggressive for local models and the previous default silently failed for anyone running vision locally. Fixes user report from GamerGB1988.

* docs: unify hooks documentation — add plugin hooks to hooks page, add session:end event The hooks page only documented gateway event hooks (HOOK.yaml system). The plugins page listed plugin hooks (pre_tool_call, etc.) that weren't referenced from the hooks page, which was confusing. Changes: - hooks.md: Add overview table showing both hook systems - hooks.md: Add Plugin Hooks section with available hooks, callback signatures, and example - hooks.md: Add missing session:end gateway event (emitted but undocumented) - hooks.md: Mark pre_llm_call, post_llm_call, on_session_start, on_session_end as planned (defined in VALID_HOOKS but not yet invoked) - hooks.md: Update info box to cross-reference plugin hooks - hooks.md: Fix heading hierarchy (gateway content as subsections) - plugins.md: Add cross-reference to hooks page for full details - plugins.md: Mark planned hooks as (planned) * feat(session_search): add recent sessions mode when query is omitted When session_search is called without a query (or with an empty query), it now returns metadata for the most recent sessions instead of erroring. This lets the agent quickly see what was worked on recently without needing specific keywords. Returns for each session: session_id, title, source, started_at, last_active, message_count, preview (first user message). Zero LLM cost — pure DB query. Current session lineage and child delegation sessions are excluded. The agent can then keyword-search specific sessions if it needs deeper context from any of them. * docs: clarify two-mode behavior in session_search schema description * fix(compression): restore sane defaults and cap summary at 12K tokens - threshold: 0.80 → 0.50 (compress at 50%, not 80%) - target_ratio: 0.40 → 0.20, now relative to threshold not total context (20% of 50% = 10% of context as tail budget) - summary ceiling: 32K → 12K (Gemini can't output more than ~12K) - Updated DEFAULT_CONFIG, config display, example config, and tests * fix: browser_vision ignores auxiliary.vision.timeout config (#2901) * docs: unify hooks documentation — add plugin hooks to hooks page, add session:end event The hooks page only documented gateway event hooks (HOOK.yaml system). The plugins page listed plugin hooks (pre_tool_call, etc.) that weren't referenced from the hooks page, which was confusing. Changes: - hooks.md: Add overview table showing both hook systems - hooks.md: Add Plugin Hooks section with available hooks, callback signatures, and example - hooks.md: Add missing session:end gateway event (emitted but undocumented) - hooks.md: Mark pre_llm_call, post_llm_call, on_session_start, on_session_end as planned (defined in VALID_HOOKS but not yet invoked) - hooks.md: Update info box to cross-reference plugin hooks - hooks.md: Fix heading hierarchy (gateway content as subsections) - plugins.md: Add cross-reference to hooks page for full details - plugins.md: Mark planned hooks as (planned) * fix: browser_vision ignores auxiliary.vision.timeout config browser_vision called call_llm() without passing a timeout parameter, so it always used the 30-second default in auxiliary_client.py. This made vision analysis with local models (llama.cpp, ollama) impossible since they typically need more than 30s for screenshot analysis. Now browser_vision reads auxiliary.vision.timeout from config.yaml (same config key that vision_analyze already uses) and passes it through to call_llm(). Also bumped the default vision timeout from 30s to 120s in both browser_vision and vision_analyze — 30s is too aggressive for local models and the previous default silently failed for anyone running vision locally. Fixes user report from GamerGB1988. * fix(skills): agent-created skills were incorrectly treated as untrusted community content _resolve_trust_level() didn't handle 'agent-created' source, so it fell through to 'community' trust level. Community policy blocks on any caution or dangerous findings, which meant common patterns like curl with env vars, systemctl, crontab, cloudflared references etc. would block skill creation/patching. The agent-created policy row already existed in INSTALL_POLICY with permissive settings (allow caution, ask on dangerous) but was never reached. Now it is. Fixes reports of skill_manage being blocked by security scanner. * fix(cli): enhance real-time reasoning output by forcing flush of long partial lines Updated the reasoning output mechanism to emit complete lines and force-flush long partial lines, ensuring reasoning is visible in real-time even without newlines. This improves user experience during reasoning sessions. * fix: skip KawaiiSpinner when TUI handles tool progress In the interactive CLI, the agent runs with quiet_mode=True and tool_progress_callback set. The quiet_mode condition triggered KawaiiSpinner for every tool call, but the TUI was already handling progress display via the spinner widget. The KawaiiSpinner writes carriage-return animation through StdoutProxy, triggering run_in_terminal() erase/redraw cycles on every flush. These redundant cycles cause the status bar to ghost into terminal scrollback. The thinking spinner already had this guard (checks thinking_callback). This extends the same pattern to the three tool spinner creation sites: concurrent tools, delegate_task, and single tool execution.

…arch#2901) * docs: unify hooks documentation — add plugin hooks to hooks page, add session:end event The hooks page only documented gateway event hooks (HOOK.yaml system). The plugins page listed plugin hooks (pre_tool_call, etc.) that weren't referenced from the hooks page, which was confusing. Changes: - hooks.md: Add overview table showing both hook systems - hooks.md: Add Plugin Hooks section with available hooks, callback signatures, and example - hooks.md: Add missing session:end gateway event (emitted but undocumented) - hooks.md: Mark pre_llm_call, post_llm_call, on_session_start, on_session_end as planned (defined in VALID_HOOKS but not yet invoked) - hooks.md: Update info box to cross-reference plugin hooks - hooks.md: Fix heading hierarchy (gateway content as subsections) - plugins.md: Add cross-reference to hooks page for full details - plugins.md: Mark planned hooks as (planned) * fix: browser_vision ignores auxiliary.vision.timeout config browser_vision called call_llm() without passing a timeout parameter, so it always used the 30-second default in auxiliary_client.py. This made vision analysis with local models (llama.cpp, ollama) impossible since they typically need more than 30s for screenshot analysis. Now browser_vision reads auxiliary.vision.timeout from config.yaml (same config key that vision_analyze already uses) and passes it through to call_llm(). Also bumped the default vision timeout from 30s to 120s in both browser_vision and vision_analyze — 30s is too aggressive for local models and the previous default silently failed for anyone running vision locally. Fixes user report from GamerGB1988.

…2973) * docs: unify hooks documentation — add plugin hooks to hooks page, add session:end event The hooks page only documented gateway event hooks (HOOK.yaml system). The plugins page listed plugin hooks (pre_tool_call, etc.) that weren't referenced from the hooks page, which was confusing. Changes: - hooks.md: Add overview table showing both hook systems - hooks.md: Add Plugin Hooks section with available hooks, callback signatures, and example - hooks.md: Add missing session:end gateway event (emitted but undocumented) - hooks.md: Mark pre_llm_call, post_llm_call, on_session_start, on_session_end as planned (defined in VALID_HOOKS but not yet invoked) - hooks.md: Update info box to cross-reference plugin hooks - hooks.md: Fix heading hierarchy (gateway content as subsections) - plugins.md: Add cross-reference to hooks page for full details - plugins.md: Mark planned hooks as (planned) * feat(session_search): add recent sessions mode when query is omitted When session_search is called without a query (or with an empty query), it now returns metadata for the most recent sessions instead of erroring. This lets the agent quickly see what was worked on recently without needing specific keywords. Returns for each session: session_id, title, source, started_at, last_active, message_count, preview (first user message). Zero LLM cost — pure DB query. Current session lineage and child delegation sessions are excluded. The agent can then keyword-search specific sessions if it needs deeper context from any of them. * docs: clarify two-mode behavior in session_search schema description * fix(compression): restore sane defaults and cap summary at 12K tokens - threshold: 0.80 → 0.50 (compress at 50%, not 80%) - target_ratio: 0.40 → 0.20, now relative to threshold not total context (20% of 50% = 10% of context as tail budget) - summary ceiling: 32K → 12K (Gemini can't output more than ~12K) - Updated DEFAULT_CONFIG, config display, example config, and tests * fix: browser_vision ignores auxiliary.vision.timeout config (NousResearch#2901) * docs: unify hooks documentation — add plugin hooks to hooks page, add session:end event The hooks page only documented gateway event hooks (HOOK.yaml system). The plugins page listed plugin hooks (pre_tool_call, etc.) that weren't referenced from the hooks page, which was confusing. Changes: - hooks.md: Add overview table showing both hook systems - hooks.md: Add Plugin Hooks section with available hooks, callback signatures, and example - hooks.md: Add missing session:end gateway event (emitted but undocumented) - hooks.md: Mark pre_llm_call, post_llm_call, on_session_start, on_session_end as planned (defined in VALID_HOOKS but not yet invoked) - hooks.md: Update info box to cross-reference plugin hooks - hooks.md: Fix heading hierarchy (gateway content as subsections) - plugins.md: Add cross-reference to hooks page for full details - plugins.md: Mark planned hooks as (planned) * fix: browser_vision ignores auxiliary.vision.timeout config browser_vision called call_llm() without passing a timeout parameter, so it always used the 30-second default in auxiliary_client.py. This made vision analysis with local models (llama.cpp, ollama) impossible since they typically need more than 30s for screenshot analysis. Now browser_vision reads auxiliary.vision.timeout from config.yaml (same config key that vision_analyze already uses) and passes it through to call_llm(). Also bumped the default vision timeout from 30s to 120s in both browser_vision and vision_analyze — 30s is too aggressive for local models and the previous default silently failed for anyone running vision locally. Fixes user report from GamerGB1988. * fix(skills): agent-created skills were incorrectly treated as untrusted community content _resolve_trust_level() didn't handle 'agent-created' source, so it fell through to 'community' trust level. Community policy blocks on any caution or dangerous findings, which meant common patterns like curl with env vars, systemctl, crontab, cloudflared references etc. would block skill creation/patching. The agent-created policy row already existed in INSTALL_POLICY with permissive settings (allow caution, ask on dangerous) but was never reached. Now it is. Fixes reports of skill_manage being blocked by security scanner. * fix(cli): enhance real-time reasoning output by forcing flush of long partial lines Updated the reasoning output mechanism to emit complete lines and force-flush long partial lines, ensuring reasoning is visible in real-time even without newlines. This improves user experience during reasoning sessions. * fix: skip KawaiiSpinner when TUI handles tool progress In the interactive CLI, the agent runs with quiet_mode=True and tool_progress_callback set. The quiet_mode condition triggered KawaiiSpinner for every tool call, but the TUI was already handling progress display via the spinner widget. The KawaiiSpinner writes carriage-return animation through StdoutProxy, triggering run_in_terminal() erase/redraw cycles on every flush. These redundant cycles cause the status bar to ghost into terminal scrollback. The thinking spinner already had this guard (checks thinking_callback). This extends the same pattern to the three tool spinner creation sites: concurrent tools, delegate_task, and single tool execution.