feat(agent): configurable timeouts for auxiliary LLM calls via config.yaml (salvage #3406) by teknium1 · Pull Request #3597 · NousResearch/hermes-agent

teknium1 · 2026-03-28T21:35:21Z

Summary

Adds per-task configurable timeouts for auxiliary LLM calls via config.yaml under auxiliary.{task}.timeout. Users with slow local models (Ollama, llama.cpp, oMLX) can now increase timeouts to prevent failures when auxiliary requests queue behind main generation.

Based on #3406 by @alanfwilliams — converted from env vars to config.yaml per project conventions, with authorship preserved.

Changes

`hermes_cli/config.py`

Added timeout to all auxiliary config sections:

auxiliary.compression.timeout: 120 (was hardcoded 45s — compression summarizes large contexts)
auxiliary.vision.timeout: 30 (already existed)
All other aux tasks: 30 (web_extract, session_search, skills_hub, approval, mcp, flush_memories)

`agent/auxiliary_client.py`

Added _get_task_timeout(task, default) helper — reads auxiliary.{task}.timeout from config
call_llm() and async_call_llm() now default timeout=None — when not explicitly passed, auto-resolves from config via the helper
Callers can still pass an explicit timeout=X to override

`agent/context_compressor.py`

Removed hardcoded timeout: 45.0 — now resolved from auxiliary.compression.timeout config (default 120s)

`agent/title_generator.py`

Bumped default from 15s to 30s (generous for local model queuing, still reasonable for a 30-token request)

Config example

auxiliary:
  compression:
    timeout: 180    # slow local model? increase this
  vision:
    timeout: 60

Testing

1082 agent + hermes_cli tests pass
No config version bump needed (deep_merge carries new keys automatically)

….yaml Add per-task timeout settings under auxiliary.{task}.timeout in config.yaml instead of hardcoded values. Users with slow local models (Ollama, llama.cpp) can now increase timeouts for compression, vision, session search, etc. Defaults: - auxiliary.compression.timeout: 120s (was hardcoded 45s) - auxiliary.vision.timeout: 30s (unchanged) - all other aux tasks: 30s (was hardcoded 30s) - title_generator: 30s (was hardcoded 15s) call_llm/async_call_llm now auto-resolve timeout from config when not explicitly passed. Callers can still override with an explicit timeout arg. Based on PR #3406 by alanfwilliams. Converted from env vars to config.yaml per project conventions.

… pages Fixes found by auditing docs against recent PRs/commits: Critical (misleading): - hooks.md: Remove stale 'planned — not yet wired' markers for 4 hooks that are now active (#3542). Add correct callback signatures. - security.md: Update tirith verdict behavior — block verdicts now go through approval flow instead of hard-blocking (#3428). Add pkill/killall self-termination guard and gateway-run backgrounding patterns (#3593). New feature docs: - configuration.md: Add tool_use_enforcement section with value table (auto/true/false/list) from #3551/#3528. - configuration.md: Expand auxiliary config with per-task timeouts (compression 120s, web_extract 30s, approval 30s) from #3597. - api-server.md: Add /v1/health alias, Security Headers section, CORS details (Max-Age, SSE headers, Idempotency-Key) from #3572/#3573/#3576/#3580/#3530. Stale/incomplete: - configuration.md: Fix Alibaba model name qwen-plus -> qwen3.5-plus (#3484). - environment-variables.md: Specify actual DashScope default URL. - cli-commands.md: Add alibaba to --provider list. - fallback-providers.md: Add Alibaba/DashScope to provider table. - email.md: Document noreply/automated sender filtering (#3606). - toolsets-reference.md: Add 4 missing platform toolsets — matrix, mattermost, dingtalk, api-server (#3583). - skills.md: List default GitHub taps including garrytan/gstack (#3605).

… pages (#3618) Fixes found by auditing docs against recent PRs/commits: Critical (misleading): - hooks.md: Remove stale 'planned — not yet wired' markers for 4 hooks that are now active (#3542). Add correct callback signatures. - security.md: Update tirith verdict behavior — block verdicts now go through approval flow instead of hard-blocking (#3428). Add pkill/killall self-termination guard and gateway-run backgrounding patterns (#3593). New feature docs: - configuration.md: Add tool_use_enforcement section with value table (auto/true/false/list) from #3551/#3528. - configuration.md: Expand auxiliary config with per-task timeouts (compression 120s, web_extract 30s, approval 30s) from #3597. - api-server.md: Add /v1/health alias, Security Headers section, CORS details (Max-Age, SSE headers, Idempotency-Key) from #3572/#3573/#3576/#3580/#3530. Stale/incomplete: - configuration.md: Fix Alibaba model name qwen-plus -> qwen3.5-plus (#3484). - environment-variables.md: Specify actual DashScope default URL. - cli-commands.md: Add alibaba to --provider list. - fallback-providers.md: Add Alibaba/DashScope to provider table. - email.md: Document noreply/automated sender filtering (#3606). - toolsets-reference.md: Add 4 missing platform toolsets — matrix, mattermost, dingtalk, api-server (#3583). - skills.md: List default GitHub taps including garrytan/gstack (#3605).

….yaml (NousResearch#3597) Add per-task timeout settings under auxiliary.{task}.timeout in config.yaml instead of hardcoded values. Users with slow local models (Ollama, llama.cpp) can now increase timeouts for compression, vision, session search, etc. Defaults: - auxiliary.compression.timeout: 120s (was hardcoded 45s) - auxiliary.vision.timeout: 30s (unchanged) - all other aux tasks: 30s (was hardcoded 30s) - title_generator: 30s (was hardcoded 15s) call_llm/async_call_llm now auto-resolve timeout from config when not explicitly passed. Callers can still override with an explicit timeout arg. Based on PR NousResearch#3406 by alanfwilliams. Converted from env vars to config.yaml per project conventions. Co-authored-by: alanfwilliams <alanfwilliams@users.noreply.github.com>

… pages (NousResearch#3618) Fixes found by auditing docs against recent PRs/commits: Critical (misleading): - hooks.md: Remove stale 'planned — not yet wired' markers for 4 hooks that are now active (NousResearch#3542). Add correct callback signatures. - security.md: Update tirith verdict behavior — block verdicts now go through approval flow instead of hard-blocking (NousResearch#3428). Add pkill/killall self-termination guard and gateway-run backgrounding patterns (NousResearch#3593). New feature docs: - configuration.md: Add tool_use_enforcement section with value table (auto/true/false/list) from NousResearch#3551/NousResearch#3528. - configuration.md: Expand auxiliary config with per-task timeouts (compression 120s, web_extract 30s, approval 30s) from NousResearch#3597. - api-server.md: Add /v1/health alias, Security Headers section, CORS details (Max-Age, SSE headers, Idempotency-Key) from NousResearch#3572/NousResearch#3573/NousResearch#3576/NousResearch#3580/NousResearch#3530. Stale/incomplete: - configuration.md: Fix Alibaba model name qwen-plus -> qwen3.5-plus (NousResearch#3484). - environment-variables.md: Specify actual DashScope default URL. - cli-commands.md: Add alibaba to --provider list. - fallback-providers.md: Add Alibaba/DashScope to provider table. - email.md: Document noreply/automated sender filtering (NousResearch#3606). - toolsets-reference.md: Add 4 missing platform toolsets — matrix, mattermost, dingtalk, api-server (NousResearch#3583). - skills.md: List default GitHub taps including garrytan/gstack (NousResearch#3605).