hermes-agent/tools
Teknium e97c0cb578
fix: replace hardcoded ~/.hermes paths with get_hermes_home() for profile support
* feat: GPT tool-use steering + strip budget warnings from history

Two changes to improve tool reliability, especially for OpenAI GPT models:

1. GPT tool-use enforcement prompt: Adds GPT_TOOL_USE_GUIDANCE to the
   system prompt when the model name contains 'gpt' and tools are loaded.
   This addresses a known behavioral pattern where GPT models describe
   intended actions ('I will run the tests') instead of actually making
   tool calls. Inspired by similar steering in OpenCode (beast.txt) and
   Cline (GPT-5.1 variant).

2. Budget warning history stripping: Budget pressure warnings injected by
   _get_budget_warning() into tool results are now stripped when
   conversation history is replayed via run_conversation(). Previously,
   these turn-scoped signals persisted across turns, causing models to
   avoid tool calls in all subsequent messages after any turn that hit
   the 70-90% iteration threshold.

* fix: replace hardcoded ~/.hermes paths with get_hermes_home() for profile support

Prep for the upcoming profiles feature — each profile is a separate
HERMES_HOME directory, so all paths must respect the env var.

Fixes:
- gateway/platforms/matrix.py: Matrix E2EE store was hardcoded to
  ~/.hermes/matrix/store, ignoring HERMES_HOME. Now uses
  get_hermes_home() so each profile gets its own Matrix state.

- gateway/platforms/telegram.py: Two locations reading config.yaml via
  Path.home()/.hermes instead of get_hermes_home(). DM topic thread_id
  persistence and hot-reload would read the wrong config in a profile.

- tools/file_tools.py: Security path for hub index blocking was
  hardcoded to ~/.hermes, would miss the actual profile's hub cache.

- hermes_cli/gateway.py: Service naming now uses the profile name
  (hermes-gateway-coder) instead of a cryptic hash suffix. Extracted
  _profile_suffix() helper shared by systemd and launchd.

- hermes_cli/gateway.py: Launchd plist path and Label now scoped per
  profile (ai.hermes.gateway-coder.plist). Previously all profiles
  would collide on the same plist file on macOS.

- hermes_cli/gateway.py: Launchd plist now includes HERMES_HOME in
  EnvironmentVariables — was missing entirely, making custom
  HERMES_HOME broken on macOS launchd (pre-existing bug).

- All launchctl commands in gateway.py, main.py, status.py updated
  to use get_launchd_label() instead of hardcoded string.

Test fixes: DM topic tests now set HERMES_HOME env var alongside
Path.home() mock. Launchd test uses get_launchd_label() for expected
commands.
2026-03-28 13:51:08 -07:00
..
browser_providers feat(browser): multi-provider cloud browser support + Browser Use integration 2026-03-17 00:16:34 -07:00
environments refactor: replace swe-rex with native Modal SDK for Modal backend (#3538) 2026-03-28 11:21:44 -07:00
neutts_samples refactor(tts): replace NeuTTS optional skill with built-in provider + setup flow 2026-03-17 02:33:12 -07:00
__init__.py chore: remove all remaining mini-swe-agent references 2026-03-24 08:19:23 -07:00
ansi_strip.py fix: strip ANSI at the source — clean terminal output before it reaches the model 2026-03-23 07:43:12 -07:00
approval.py fix: make tirith block verdicts approvable instead of hard-blocking (#3428) 2026-03-27 13:22:01 -07:00
browser_tool.py chore: fix 154 f-strings, simplify getattr/URL patterns, remove dead code (#3119) 2026-03-25 19:47:58 -07:00
checkpoint_manager.py chore: fix 154 f-strings, simplify getattr/URL patterns, remove dead code (#3119) 2026-03-25 19:47:58 -07:00
clarify_tool.py chore: remove ~100 unused imports across 55 files (#3016) 2026-03-25 15:02:03 -07:00
code_execution_tool.py feat: env var passthrough for skills and user config (#2807) 2026-03-24 08:19:34 -07:00
cronjob_tools.py fix: normalize repeat<=0 to None to prevent cron jobs deleting after first run (#2612) 2026-03-23 06:35:43 -07:00
debug_helpers.py refactor: consolidate debug logging across tools with shared DebugSession class 2026-02-21 03:53:24 -08:00
delegate_tool.py fix(security): restrict subagent toolsets to parent's enabled set (#3269) 2026-03-26 14:50:26 -07:00
env_passthrough.py feat: env var passthrough for skills and user config (#2807) 2026-03-24 08:19:34 -07:00
file_operations.py chore: remove ~100 unused imports across 55 files (#3016) 2026-03-25 15:02:03 -07:00
file_tools.py fix: replace hardcoded ~/.hermes paths with get_hermes_home() for profile support 2026-03-28 13:51:08 -07:00
fuzzy_match.py chore: fix 154 f-strings, simplify getattr/URL patterns, remove dead code (#3119) 2026-03-25 19:47:58 -07:00
homeassistant_tool.py feat(tools): centralize tool emoji metadata in registry + skin integration 2026-03-15 20:21:21 -07:00
honcho_tools.py fix(honcho): isolate session routing for multi-user gateway (#1500) 2026-03-16 00:23:47 -07:00
image_generation_tool.py chore: remove ~100 unused imports across 55 files (#3016) 2026-03-25 15:02:03 -07:00
interrupt.py feat: enhance interrupt handling and container resource configuration 2026-02-23 02:11:33 -08:00
mcp_oauth.py chore: fix 154 f-strings, simplify getattr/URL patterns, remove dead code (#3119) 2026-03-25 19:47:58 -07:00
mcp_tool.py fix: guard config.get() against YAML null values to prevent AttributeError (#3377) 2026-03-27 04:03:00 -07:00
memory_tool.py refactor: consolidate get_hermes_home() and parse_reasoning_effort() (#3062) 2026-03-25 15:54:28 -07:00
mixture_of_agents_tool.py fix: guard aux LLM calls against None content + reasoning fallback + retry (salvage #3389) (#3449) 2026-03-27 15:28:19 -07:00
neutts_synth.py fix(tts): document NeuTTS provider and align install guidance (#1903) 2026-03-18 02:55:30 -07:00
openrouter_client.py refactor: route ad-hoc LLM consumers through centralized provider router 2026-03-11 20:02:36 -07:00
patch_parser.py fix: handle addition-only hunks in V4A patch parser (#3325) 2026-03-26 19:38:04 -07:00
process_registry.py chore: remove ~100 unused imports across 55 files (#3016) 2026-03-25 15:02:03 -07:00
registry.py perf(ttft): salvage easy-win startup optimizations from #3346 (#3395) 2026-03-27 07:49:44 -07:00
rl_training_tool.py chore: fix 154 f-strings, simplify getattr/URL patterns, remove dead code (#3119) 2026-03-25 19:47:58 -07:00
send_message_tool.py fix: add request timeouts to send_message_tool HTTP calls (#3162) 2026-03-26 01:58:11 -07:00
session_search_tool.py fix: session_search fallback preview on summarization failure (salvage #3413) (#3478) 2026-03-27 21:27:51 -07:00
skill_manager_tool.py perf(ttft): cache skills prompt with shared skill_utils module (salvage #3366) (#3421) 2026-03-27 10:54:02 -07:00
skills_guard.py fix: guard aux LLM calls against None content + reasoning fallback + retry (salvage #3389) (#3449) 2026-03-27 15:28:19 -07:00
skills_hub.py fix(skills): preserve trust for skills-sh identifiers + reduce resolution churn (#3251) 2026-03-26 13:40:21 -07:00
skills_sync.py refactor: consolidate get_hermes_home() and parse_reasoning_effort() (#3062) 2026-03-25 15:54:28 -07:00
skills_tool.py perf(ttft): cache skills prompt with shared skill_utils module (salvage #3366) (#3421) 2026-03-27 10:54:02 -07:00
terminal_tool.py refactor: replace swe-rex with native Modal SDK for Modal backend (#3538) 2026-03-28 11:21:44 -07:00
tirith_security.py chore: fix 154 f-strings, simplify getattr/URL patterns, remove dead code (#3119) 2026-03-25 19:47:58 -07:00
todo_tool.py feat(tools): centralize tool emoji metadata in registry + skin integration 2026-03-15 20:21:21 -07:00
transcription_tools.py refactor: consolidate get_hermes_home() and parse_reasoning_effort() (#3062) 2026-03-25 15:54:28 -07:00
tts_tool.py fix: guard config.get() against YAML null values to prevent AttributeError (#3377) 2026-03-27 04:03:00 -07:00
url_safety.py fix(security): add SSRF protection to vision_tools and web_tools (hardened) 2026-03-23 15:40:42 -07:00
vision_tools.py fix: guard aux LLM calls against None content + reasoning fallback + retry (salvage #3389) (#3449) 2026-03-27 15:28:19 -07:00
voice_mode.py chore: remove ~100 unused imports across 55 files (#3016) 2026-03-25 15:02:03 -07:00
web_tools.py fix: guard aux LLM calls against None content + reasoning fallback + retry (salvage #3389) (#3449) 2026-03-27 15:28:19 -07:00
website_policy.py refactor: consolidate get_hermes_home() and parse_reasoning_effort() (#3062) 2026-03-25 15:54:28 -07:00