hermes-agent

name	hermes-agent
description	Complete guide to using and extending Hermes Agent — CLI usage, setup, configuration, spawning additional agents, gateway platforms, skills, voice, tools, profiles, and a concise contributor reference. Load this skill when helping users configure Hermes, troubleshoot issues, spawn agent instances, or make code contributions.
version	2.0.0
author	Hermes Agent + Teknium
license	MIT
metadata	{"hermes":{"tags":["hermes","setup","configuration","multi-agent","spawning","gateway","tools","skills","voice","cli","development","routing","s6","container-supervision"],"related_skills":["claude-code","opencode"]}}

Hermes Agent

Consolidated Hermes internals

This umbrella also owns former narrow Hermes maintenance skills:

• hermes-routing: configuring skill-router/project-router/skill-index mappings, especially Chinese trigger aliases and route misses.
• hermes-s6-container-supervision: modifying or debugging the Docker image s6-overlay supervision tree, gateway/profile services, and Architecture B main-program patterns.

Full original packages are preserved under references/absorbed-skills/ for detailed recipes and support files.

Hermes Agent is an open-source AI agent framework by Nous Research that runs in your terminal, messaging platforms, and IDEs. It belongs to the same category as Claude Code (Anthropic), Codex (OpenAI), and OpenClaw — autonomous coding and task-execution agents that use tool calling to interact with your system. Hermes works with any LLM provider (OpenRouter, Anthropic, OpenAI, DeepSeek, local models, and 15+ others) and runs on Linux, macOS, and WSL.

What makes Hermes different:

• Self-improving through skills — Hermes learns from experience by saving reusable procedures as skills. When it solves a complex problem, discovers a workflow, or gets corrected, it can persist that knowledge as a skill document that loads into future sessions. Skills accumulate over time, making the agent better at your specific tasks and environment.
• Persistent memory across sessions — remembers who you are, your preferences, environment details, and lessons learned. Pluggable memory backends (built-in, Honcho, Mem0, and more) let you choose how memory works.
• Multi-platform gateway — the same agent runs on Telegram, Discord, Slack, WhatsApp, Signal, Matrix, Email, and 8+ other platforms with full tool access, not just chat.
• Provider-agnostic — swap models and providers mid-workflow without changing anything else. Credential pools rotate across multiple API keys automatically.
• Profiles — run multiple independent Hermes instances with isolated configs, sessions, skills, and memory.
• Extensible — plugins, MCP servers, custom tools, webhook triggers, cron scheduling, and the full Python ecosystem.

People use Hermes for software development, research, system administration, data analysis, content creation, home automation, and anything else that benefits from an AI agent with persistent context and full system access.

This skill helps you work with Hermes Agent effectively — setting it up, configuring features, spawning additional agent instances, troubleshooting issues, finding the right commands and settings, and understanding how the system works when you need to extend or contribute to it.

Docs: https://hermes-agent.nousresearch.com/docs/

Quick Start

# Install
curl -fsSL https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.sh | bash

# Interactive chat (default)
hermes

# Single query
hermes chat -q "What is the capital of France?"

# Setup wizard
hermes setup

# Change model/provider
hermes model

# Check health
hermes doctor

---

CLI Reference

Global Flags

hermes [flags] [command]

  --version, -V             Show version
  --resume, -r SESSION      Resume session by ID or title
  --continue, -c [NAME]     Resume by name, or most recent session
  --worktree, -w            Isolated git worktree mode (parallel agents)
  --skills, -s SKILL        Preload skills (comma-separate or repeat)
  --profile, -p NAME        Use a named profile
  --yolo                    Skip dangerous command approval
  --pass-session-id         Include session ID in system prompt

No subcommand defaults to chat.

Chat

hermes chat [flags]
  -q, --query TEXT          Single query, non-interactive
  -m, --model MODEL         Model (e.g. anthropic/claude-sonnet-4)
  -t, --toolsets LIST       Comma-separated toolsets
  --provider PROVIDER       Force provider (openrouter, anthropic, nous, etc.)
  -v, --verbose             Verbose output
  -Q, --quiet               Suppress banner, spinner, tool previews
  --checkpoints             Enable filesystem checkpoints (/rollback)
  --source TAG              Session source tag (default: cli)

Configuration

hermes setup [section]      Interactive wizard (model|terminal|gateway|tools|agent)
hermes model                Interactive model/provider picker
hermes config               View current config
hermes config edit          Open config.yaml in $EDITOR
hermes config set KEY VAL   Set a config value
hermes config path          Print config.yaml path
hermes config env-path      Print .env path
hermes config check         Check for missing/outdated config
hermes config migrate       Update config with new options
hermes login [--provider P] OAuth login (nous, openai-codex)
hermes logout               Clear stored auth
hermes doctor [--fix]       Check dependencies and config
hermes status [--all]       Show component status

Tools & Skills

hermes tools                Interactive tool enable/disable (curses UI)
hermes tools list           Show all tools and status
hermes tools enable NAME    Enable a toolset
hermes tools disable NAME   Disable a toolset

hermes skills list          List installed skills
hermes skills search QUERY  Search the skills hub
hermes skills install ID    Install a skill
hermes skills inspect ID    Preview without installing
hermes skills config        Enable/disable skills per platform
hermes skills check         Check for updates
hermes skills update        Update outdated skills
hermes skills uninstall N   Remove a hub skill
hermes skills publish PATH  Publish to registry
hermes skills browse        Browse all available skills
hermes skills tap add REPO  Add a GitHub repo as skill source

MCP Servers

hermes mcp serve            Run Hermes as an MCP server
hermes mcp add NAME         Add an MCP server (--url or --command)
hermes mcp remove NAME      Remove an MCP server
hermes mcp list             List configured servers
hermes mcp test NAME        Test connection
hermes mcp configure NAME   Toggle tool selection

Gateway (Messaging Platforms)

hermes gateway run          Start gateway foreground
hermes gateway install      Install as background service
hermes gateway start/stop   Control the service
hermes gateway restart      Restart the service
hermes gateway status       Check status
hermes gateway setup        Configure platforms

Supported platforms: Telegram, Discord, Slack, WhatsApp, Signal, Email, SMS, Matrix, Mattermost, Home Assistant, DingTalk, Feishu, WeCom, API Server, Webhooks, Open WebUI.

Platform docs: https://hermes-agent.nousresearch.com/docs/user-guide/messaging/

Sessions

hermes sessions list        List recent sessions
hermes sessions browse      Interactive picker
hermes sessions export OUT  Export to JSONL
hermes sessions rename ID T Rename a session
hermes sessions delete ID   Delete a session
hermes sessions prune       Clean up old sessions (--older-than N days)
hermes sessions stats       Session store statistics

Cron Jobs

hermes cron list            List jobs (--all for disabled)
hermes cron create SCHED    Create: '30m', 'every 2h', '0 9 * * *'
hermes cron edit ID         Edit schedule, prompt, delivery
hermes cron pause/resume ID Control job state
hermes cron run ID          Trigger on next tick
hermes cron remove ID       Delete a job
hermes cron status          Scheduler status

Webhooks

hermes webhook subscribe N  Create route at /webhooks/<name>
hermes webhook list         List subscriptions
hermes webhook remove NAME  Remove a subscription
hermes webhook test NAME    Send a test POST

Profiles

hermes profile list         List all profiles
hermes profile create NAME  Create (--clone, --clone-all, --clone-from)
hermes profile use NAME     Set sticky default
hermes profile delete NAME  Delete a profile
hermes profile show NAME    Show details
hermes profile alias NAME   Manage wrapper scripts
hermes profile rename A B   Rename a profile
hermes profile export NAME  Export to tar.gz
hermes profile import FILE  Import from archive

Credential Pools

hermes auth add             Interactive credential wizard
hermes auth list [PROVIDER] List pooled credentials
hermes auth remove P INDEX  Remove by provider + index
hermes auth reset PROVIDER  Clear exhaustion status

Bulk custom-provider key insertion: see references/custom-provider-credential-pool-bulk-keys.md for the safe auth.json credential_pool merge pattern, including backup, manual source entries, and no-secret verification.

Bulk CPA Manager OpenAI-compatible provider key insertion and paced key-pool maintenance: see references/cpa-openai-compatible-provider-bulk-keys.md for the /v0/management/openai-compatibility API pattern, management-key auth, scheduler cooldown policy, backup, merge, and no-secret verification.

Other

hermes insights [--days N]  Usage analytics
hermes update               Update to latest version
hermes pairing list/approve/revoke  DM authorization
hermes plugins list/install/remove  Plugin management
hermes honcho setup/status  Honcho memory integration
hermes memory setup/status/off  Memory provider config
hermes completion bash|zsh  Shell completions
hermes acp                  ACP server (IDE integration)
hermes claw migrate         Migrate from OpenClaw
hermes uninstall            Uninstall Hermes

---

Slash Commands (In-Session)

Type these during an interactive chat session.

Session Control

/new (/reset)        Fresh session
/clear               Clear screen + new session (CLI)
/retry               Resend last message
/undo                Remove last exchange
/title [name]        Name the session
/compress            Manually compress context
/stop                Kill background processes
/rollback [N]        Restore filesystem checkpoint
/background <prompt> Run prompt in background
/queue <prompt>      Queue for next turn
/resume [name]       Resume a named session

Configuration

/config              Show config (CLI)
/model [name]        Show or change model
/provider            Show provider info
/prompt [text]       View/set system prompt (CLI)
/personality [name]  Set personality
/reasoning [level]   Set reasoning (none|low|medium|high|xhigh|show|hide)
/verbose             Cycle: off → new → all → verbose
/voice [on|off|tts]  Voice mode
/yolo                Toggle approval bypass
/skin [name]         Change theme (CLI)
/statusbar           Toggle status bar (CLI)

Tools & Skills

/tools               Manage tools (CLI)
/toolsets            List toolsets (CLI)
/skills              Search/install skills (CLI)
/skill <name>        Load a skill into session
/cron                Manage cron jobs (CLI)
/reload-mcp          Reload MCP servers
/plugins             List plugins (CLI)

Info

/help                Show commands
/commands [page]     Browse all commands (gateway)
/usage               Token usage
/insights [days]     Usage analytics
/status              Session info (gateway)
/profile             Active profile info

Exit

/quit (/exit, /q)    Exit CLI

---

Key Paths & Config

~/.hermes/config.yaml       Main configuration
~/.hermes/.env              API keys and secrets
~/.hermes/skills/           Installed skills
~/.hermes/sessions/         Session transcripts
~/.hermes/logs/             Gateway and error logs
~/.hermes/auth.json         OAuth tokens and credential pools
~/.hermes/hermes-agent/     Source code (if git-installed)

Profiles use ~/.hermes/profiles/<name>/ with the same layout.

Config Sections

Edit with hermes config edit or hermes config set section.key value.

Section	Key options
model	default, provider, base_url, api_key, context_length
agent	max_turns (90), tool_use_enforcement
terminal	backend (local/docker/ssh/modal), cwd, timeout (180)
compression	enabled, threshold (0.50), target_ratio (0.20)
display	skin, tool_progress, show_reasoning, show_cost
stt	enabled, provider (local/groq/openai)
tts	provider (edge/elevenlabs/openai/kokoro/fish)
memory	memory_enabled, user_profile_enabled, provider
security	tirith_enabled, website_blocklist
delegation	model, provider, max_iterations (50)
smart_model_routing	enabled, cheap_model
checkpoints	enabled, max_snapshots (50)

Full config reference: https://hermes-agent.nousresearch.com/docs/user-guide/configuration

Providers

18 providers supported. Set via hermes model or hermes setup.

Provider	Auth	Key env var
OpenRouter	API key	OPENROUTER_API_KEY
Anthropic	API key	ANTHROPIC_API_KEY
Nous Portal	OAuth	hermes login --provider nous
OpenAI Codex	OAuth	hermes login --provider openai-codex
GitHub Copilot	Token	COPILOT_GITHUB_TOKEN
DeepSeek	API key	DEEPSEEK_API_KEY
Hugging Face	Token	HF_TOKEN
Z.AI / GLM	API key	GLM_API_KEY
MiniMax	API key	MINIMAX_API_KEY
Kimi / Moonshot	API key	KIMI_API_KEY
Alibaba / DashScope	API key	DASHSCOPE_API_KEY
Kilo Code	API key	KILOCODE_API_KEY
Custom endpoint	Config	model.base_url + model.api_key in config.yaml

Plus: AI Gateway, OpenCode Zen, OpenCode Go, MiniMax CN, GitHub Copilot ACP.

Full provider docs: https://hermes-agent.nousresearch.com/docs/integrations/providers

Toolsets

Enable/disable via hermes tools (interactive) or hermes tools enable/disable NAME.

Toolset	What it provides
web	Web search and content extraction
browser	Browser automation (Browserbase, Camofox, or local Chromium)
terminal	Shell commands and process management
file	File read/write/search/patch
code_execution	Sandboxed Python execution
vision	Image analysis
image_gen	AI image generation
tts	Text-to-speech
skills	Skill browsing and management
memory	Persistent cross-session memory
session_search	Search past conversations
delegation	Subagent task delegation
cronjob	Scheduled task management
clarify	Ask user clarifying questions
moa	Mixture of Agents (off by default)
homeassistant	Smart home control (off by default)

Tool changes take effect on /reset (new session). They do NOT apply mid-conversation to preserve prompt caching.

---

Voice & Transcription

STT (Voice → Text)

Voice messages from messaging platforms are auto-transcribed.

Provider priority (auto-detected):

1. Local faster-whisper — free, no API key: pip install faster-whisper
2. Groq Whisper — free tier: set GROQ_API_KEY
3. OpenAI Whisper — paid: set VOICE_TOOLS_OPENAI_KEY

Config:

stt:
  enabled: true
  provider: local        # local, groq, openai
  local:
    model: base          # tiny, base, small, medium, large-v3

TTS (Text → Voice)

Provider	Env var	Free?
Edge TTS	None	Yes (default)
ElevenLabs	ELEVENLABS_API_KEY	Free tier
OpenAI	VOICE_TOOLS_OPENAI_KEY	Paid
Kokoro (local)	None	Free
Fish Audio	FISH_AUDIO_API_KEY	Free tier

Voice commands: /voice on (voice-to-voice), /voice tts (always voice), /voice off.

---

Spawning Additional Hermes Instances

Run additional Hermes processes as fully independent subprocesses — separate sessions, tools, and environments.

When to Use This vs delegate_task

	delegate_task	Spawning hermes process
Isolation	Separate conversation, shared process	Fully independent process
Duration	Minutes (bounded by parent loop)	Hours/days
Tool access	Subset of parent's tools	Full tool access
Interactive	No	Yes (PTY mode)
Use case	Quick parallel subtasks	Long autonomous missions

One-Shot Mode

terminal(command="hermes chat -q 'Research GRPO papers and write summary to ~/research/grpo.md'", timeout=300)

# Background for long tasks:
terminal(command="hermes chat -q 'Set up CI/CD for ~/myapp'", background=true)

Interactive PTY Mode (via tmux)

Hermes uses prompt_toolkit, which requires a real terminal. Use tmux for interactive spawning:

# Start
terminal(command="tmux new-session -d -s agent1 -x 120 -y 40 'hermes'", timeout=10)

# Wait for startup, then send a message
terminal(command="sleep 8 && tmux send-keys -t agent1 'Build a FastAPI auth service' Enter", timeout=15)

# Read output
terminal(command="sleep 20 && tmux capture-pane -t agent1 -p", timeout=5)

# Send follow-up
terminal(command="tmux send-keys -t agent1 'Add rate limiting middleware' Enter", timeout=5)

# Exit
terminal(command="tmux send-keys -t agent1 '/exit' Enter && sleep 2 && tmux kill-session -t agent1", timeout=10)

Multi-Agent Coordination

# Agent A: backend
terminal(command="tmux new-session -d -s backend -x 120 -y 40 'hermes -w'", timeout=10)
terminal(command="sleep 8 && tmux send-keys -t backend 'Build REST API for user management' Enter", timeout=15)

# Agent B: frontend
terminal(command="tmux new-session -d -s frontend -x 120 -y 40 'hermes -w'", timeout=10)
terminal(command="sleep 8 && tmux send-keys -t frontend 'Build React dashboard for user management' Enter", timeout=15)

# Check progress, relay context between them
terminal(command="tmux capture-pane -t backend -p | tail -30", timeout=5)
terminal(command="tmux send-keys -t frontend 'Here is the API schema from the backend agent: ...' Enter", timeout=5)

Session Resume

# Resume most recent session
terminal(command="tmux new-session -d -s resumed 'hermes --continue'", timeout=10)

# Resume specific session
terminal(command="tmux new-session -d -s resumed 'hermes --resume 20260225_143052_a1b2c3'", timeout=10)

Tips

• Prefer delegate_task for quick subtasks — less overhead than spawning a full process
• Use -w (worktree mode) when spawning agents that edit code — prevents git conflicts
• Set timeouts for one-shot mode — complex tasks can take 5-10 minutes
• Use hermes chat -q for fire-and-forget — no PTY needed
• Use tmux for interactive sessions — raw PTY mode has \r vs \n issues with prompt_toolkit
• For scheduled tasks, use the cronjob tool instead of spawning — handles delivery and retry

---

Troubleshooting

Background process running but no output

Full fix reference: references/background-process-maintenance.md

Quick checks: inspect wrapper and child PIDs with ps/pgrep -P, use lsof for active sockets, and on macOS use sample <child-pid> 3 1 to distinguish sleep, network wait, and CPU loop. For long-running helpers, restart with python3 -u and explicit log-file writes under ~/.hermes/logs/ before every long sleep or request.

GPT model misroutes read instead of read_file (loop / hang)

Full fix reference: references/gpt-read-tool-misroute-failure.md

Quick fix: add alias mapping at gateway layer:

if tool_name == "read" and "path" in args:
    tool_name = "read_file"

Installing external skill repos (Aegis, Superpowers-style)

Full reference: references/aegis-multi-host-install.md

Pattern: clone repo → symlink into the target agent's skills dir only when explicitly requested → restart that agent. Keep Hermes routing/index as the control plane, and prefer direct runtime SKILL.md reads over mirrored host skill directories.

Voice not working

1. Check stt.enabled: true in config.yaml
2. Verify provider: pip install faster-whisper or set API key
3. Restart gateway: /restart

Tool not available

1. hermes tools — check if toolset is enabled for your platform
2. Some tools need env vars (check .env)
3. /reset after enabling tools

Third-party MCP server setup in Windsurf / VS Code-family IDEs

When a user asks to install or wire an external MCP server into Windsurf, GitHub Copilot in Windsurf, or another VS Code-family IDE, do not assume the same config schema everywhere.

Observed working pattern on macOS:

• Windsurf reads ~/.codeium/windsurf/mcp_config.json
• Windsurf expects top-level key mcpServers
• HTTP servers use serverUrl, not VS Code's servers -> type: http -> url
• Existing servers in that file should be preserved and merged, not overwritten blindly

Example Windsurf entry:

{
  "mcpServers": {
    "mcp-chat": {
      "serverUrl": "http://127.0.0.1:8080/mcp"
    }
  }
}

Contrast:

• VS Code / GitHub Copilot examples may use ~/.vscode/mcp.json with:

{
  "servers": {
    "mcp-chat": {
      "type": "http",
      "url": "http://127.0.0.1:8080/mcp"
    }
  }
}

• Do not paste this VS Code schema directly into Windsurf.

Python interpreter mismatch when launching external MCP/chat servers

If a cloned Python MCP/chat server appears to hang or not bind its port even though the process stays alive:

• verify which python3 is actually being used
• on macOS, /usr/bin/python3 may differ from Homebrew Python and lack the installed packages
• a shell may resolve python3 differently from the package-install environment

Observed failure pattern:

• background process remains running
• no port is listening initially
• no obvious logs surface
• switching from system Python to Homebrew Python fixes startup

Practical check:

which python3
python3 -c 'import mcp, websockets; print("ok")'

If needed, launch explicitly with the interpreter that has the deps, e.g. /opt/homebrew/bin/python3 server.py.

Streamable HTTP MCP probe gotcha

Some streamable HTTP MCP servers require:

• Content-Type: application/json
• Accept: application/json, text/event-stream

If you POST initialize without the combined Accept header, the server may reply:

• Not Acceptable: Client must accept both application/json and text/event-stream

Minimal probe:

curl -sS -X POST http://127.0.0.1:8080/mcp \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json, text/event-stream' \
  -d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2025-03-26","capabilities":{},"clientInfo":{"name":"probe","version":"1.0"}}}'

Use this to distinguish config/schema problems from transport/header problems.

macOS launchd for third-party local MCP/chat servers

When making a third-party local MCP/chat server persistent on macOS, prefer a per-user LaunchAgent over ad-hoc background shells.

Observed working pattern:

• plist at ~/Library/LaunchAgents/<label>.plist
• RunAtLoad + KeepAlive
• explicit interpreter path (for example /opt/homebrew/bin/python3)
• explicit WorkingDirectory
• explicit log files in ~/Library/Logs/
• if the server supports env-configured ports, set them in EnvironmentVariables

Skeleton:

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
<dict>
  <key>Label</key>
  <string>com.example.mcp-chat</string>
  <key>ProgramArguments</key>
  <array>
    <string>/opt/homebrew/bin/python3</string>
    <string>/path/to/server.py</string>
  </array>
  <key>WorkingDirectory</key>
  <string>/path/to/repo</string>
  <key>RunAtLoad</key>
  <true/>
  <key>KeepAlive</key>
  <true/>
  <key>StandardOutPath</key>
  <string>/Users/NAME/Library/Logs/mcp-chat.stdout.log</string>
  <key>StandardErrorPath</key>
  <string>/Users/NAME/Library/Logs/mcp-chat.stderr.log</string>
  <key>EnvironmentVariables</key>
  <dict>
    <key>PATH</key>
    <string>/opt/homebrew/bin:/usr/local/bin:/usr/bin:/bin:/usr/sbin:/sbin</string>
    <key>MCP_CHAT_PORT</key>
    <string>8090</string>
    <key>MCP_CHAT_WS_PORT</key>
    <string>8091</string>
  </dict>
</dict>
</plist>

Load/reload:

launchctl bootout gui/$(id -u) "$HOME/Library/LaunchAgents/com.example.mcp-chat.plist" >/dev/null 2>&1 || true
launchctl bootstrap gui/$(id -u) "$HOME/Library/LaunchAgents/com.example.mcp-chat.plist"
launchctl kickstart -k gui/$(id -u)/com.example.mcp-chat

Verify:

launchctl print gui/$(id -u)/com.example.mcp-chat | sed -n '1,120p'
lsof -nP -iTCP:8090 -sTCP:LISTEN
curl -sS http://127.0.0.1:8090/v1/models

Important pitfall:

• do not try to change the listen port only with launchd Sockets unless the app is socket-activation aware or actually consumes launchd-passed file descriptors
• many Python servers still bind the port from their own code/config, so Sockets alone can leave launchd showing service name = 8090 while the app still serves 8080
• prefer patching the app to read env vars like MCP_CHAT_PORT / MCP_CHAT_WS_PORT, then set those vars in the plist

Model/provider issues

1. hermes doctor — check config and dependencies
2. hermes login — re-authenticate OAuth providers
3. Check .env has the right API key

Changes not taking effect

• Tools/skills: /reset starts a new session with updated toolset
• Config changes: /restart reloads gateway config
• Code changes: Restart the CLI or gateway process

Skills not showing

1. hermes skills list — verify installed
2. hermes skills config — check platform enablement
3. Load explicitly: /skill name or hermes -s name

Installing local GitHub skills when hermes skills inspect/install <url> does not support the repo

Some GitHub skill repos are not indexed by Hermes skill sources, so hermes skills inspect https://github.com/org/repo can return Could not find ... in any source even when the repo is valid.

Working fallback:

git clone https://github.com/org/repo /tmp/repo-review
# review README/SKILL/install/runtime first
mkdir -p ~/.hermes/skills/<category>
cp -R /tmp/repo-review ~/.hermes/skills/<category>/<skill-name>
hermes skills list | grep -i '<skill-name>'

If the repo ships a binary installer but release download fails, prefer a documented local build path (cargo build --release, etc.) over inventing a second install script. Keep copied runtime skill and installed binary separate: ~/.hermes/skills/... for the skill, ~/.local/bin/... for the executable.

Cross-agent output style normalization

When the user corrects global response format/style, align all active agent prompt surfaces instead of fixing only the current Hermes repo file. See references/cross-agent-output-style-normalization.md for the local Codex / Claude / OpenClaw prompt files and the verification search terms. In particular, do not keep forced bracketed audit fields in global prompt files unless the user explicitly asks for that structured protocol.

Skill context budget exceeded / target skill not visible

Symptom:

• Hermes or a downstream agent reports something like Exceeded skills context budget of 2%
• the visible skill list is truncated
• the skill you need is installed locally but did not make it into the model-visible window

What it means:

• this is usually a routing/exposure problem, not a missing-install problem
• the agent cannot auto-discover every installed skill when the skill catalog is too large for the prompt budget

Fastest fixes:

1. Preload the exact skill explicitly:
   • hermes --skills <exact-route>
   • hermes --skills autonomous-ai-agents/hermes-agent,<exact-route>
2. In-session, force-load the skill with /skill <route> instead of relying on broad auto-discovery.
3. Reduce default routing exposure so fewer unrelated skills are described up front.
4. For repeated workflows, create a lighter profile with only the needed routing/skills enabled.

Where to tune:

• ~/.hermes/routing/skill-router.md
• project-local .hermes/routing/project-router.md
• ~/.hermes/routing/skill-index.json
• oversized skill descriptions/frontmatter that waste prompt budget

Practical guidance:

• if a task repeatedly needs one specific skill, prefer explicit preload over hoping the global skill index will fit
• if a downstream agent can read the Hermes routing files directly, prefer routing-first resolution over dumping a giant available-skills catalog into prompt context
• if an external host agent can read the Hermes routing files directly, prefer routing-first resolution over dumping a giant available-skills catalog into prompt context
• keep Hermes routing/index as the control plane; do not mirror Hermes skills into another host agent's skill directory unless the user explicitly asks for that shared-library behavior
• after changing routing, index triggers, or skill mappings, restart or open a fresh session because external agents may cache the previous skill list

Common durable fix:

• add hard first-load rules for Hermes/routing topics in skill-router.md
• add project-specific overrides in project-router.md
• add matching trigger phrases for aa/hermes-agent and the target runtime skill in skill-index.json
• when the downstream agent still cannot read routing files, pass the exact resolved aliases and a compact routing instruction block inline

Bulk skill renaming / shortening

When a user asks to shorten or rename many local Hermes skills at once, treat it as a compatibility-breaking library migration, not a cosmetic edit.

Observed constraints:

• Changing only frontmatter name: is insufficient; runtime lookup commonly depends on directory basename too.
• Router and governance docs can pin explicit runtime names and must be updated in the same task.
• Existing cross-agent consumers, shell snippets, and user muscle memory will break after aggressive shortening.
• A partially interrupted bulk pass can leave the skill tree in a mixed state where some directory names are shortened but some name: fields still hold older names.
• Tool surfaces that show numbered lines can contaminate copied skill files if that text is written back directly.

Safe pattern:

1. Inventory the current runtime tree under ~/.hermes/skills.
2. Derive a deterministic rename map with collision handling before mutating files.
3. Rename directory basenames and then rewrite each SKILL.md frontmatter name: to match the final basename.
4. In the same task, patch routing/governance/index docs under ~/.hermes/routing that mention the old runtime names.
5. Re-verify:
   • every SKILL.md has a name: line
   • every name: equals its containing directory basename
   • no duplicate skill names remain
   • high-value router docs no longer reference superseded names
6. If a file was rewritten from numbered tool output, strip injected N| prefixes before final verification.

Verification probe:

python3 - <<'PY'
import os, re, json, pathlib
root=os.path.expanduser('~/.hermes/skills')
seen={}; mism=[]; dups=[]; missing=[]
for dp, dn, fn in os.walk(root):
    if 'SKILL.md' not in fn:
        continue
    p=os.path.join(dp,'SKILL.md')
    txt=pathlib.Path(p).read_text()
    m=re.search(r'(?m)^name:\s*(.+?)\s*$', txt)
    if not m:
        missing.append(p)
        continue
    name=m.group(1).strip()
    base=os.path.basename(dp)
    if name != base:
        mism.append((p, name, base))
    if name in seen:
        dups.append((name, seen[name], p))
    seen[name]=p
print(json.dumps({
    'skills': len(seen),
    'missing_name': len(missing),
    'mismatches': len(mism),
    'duplicates': len(dups),
}, ensure_ascii=False))
PY

Pitfalls:

• Do not trust stale counts from prior sessions; rescan the live tree first.
• Do not stop after renaming directories; fix frontmatter and router docs in the same pass.
• Do not copy text from line-numbered readouts back into SKILL.md verbatim.
• If loaded skills disappear after the migration, first try the new shorter names rather than assuming the library is broken.

hermes update fails on dependency install

Common symptoms:

• Failed to fetch: https://pypi.org/simple/...
• tls handshake eof
• fallback says optional extras failed, then base install fails
• venv/bin/python: No module named pip
• No matching distribution found for wheel while using a mirror

Fast recovery on macOS/Linux:

cd ~/.hermes/hermes-agent
source venv/bin/activate || true
python3 -m ensurepip --upgrade
PIP_INDEX_URL=https://pypi.org/simple PIP_TRUSTED_HOST=pypi.org python3 -m pip install --upgrade pip setuptools wheel
UV_INDEX_URL=https://pypi.org/simple ~/.local/bin/uv pip install -e . --quiet
hermes --version

If the machine has a bad global pip mirror, inspect and fix it:

python3 -m pip config list -v
nl -ba ~/.config/pip/pip.conf

If index-url points at a stale or incomplete mirror (for example Tsinghua tuna) and installs are failing, replace it with:

[global]
index-url = https://pypi.org/simple

Then verify:

python3 -m pip index versions wheel | head -n 5

Notes:

• uv pip can succeed once UV_INDEX_URL is forced to PyPI even if global pip config is broken.
• Missing pip inside the Hermes venv is recoverable with python3 -m ensurepip --upgrade.
• When debugging install failures, check both network reachability to https://pypi.org/simple/... and local pip config before changing project code.

Gateway issues

Check logs first:

grep -i "failed to send\|error" ~/.hermes/logs/gateway.log | tail -20

---

Where to Find Things

Looking for...	Location
Config options	hermes config edit or Configuration docs
Available tools	hermes tools list or Tools reference
Slash commands	/help in session or Slash commands reference
Skills catalog	hermes skills browse or Skills catalog
Provider setup	hermes model or Providers guide
Platform setup	hermes gateway setup or Messaging docs
MCP servers	hermes mcp list or MCP guide
Profiles	hermes profile list or Profiles docs
Cron jobs	hermes cron list or Cron docs
Memory	hermes memory status or Memory docs
Env variables	hermes config env-path or Env vars reference
CLI commands	hermes --help or CLI reference
Gateway logs	~/.hermes/logs/gateway.log
Session files	~/.hermes/sessions/ or hermes sessions browse
Source code	~/.hermes/hermes-agent/

---

Contributor Quick Reference

For occasional contributors and PR authors. Full developer docs: https://hermes-agent.nousresearch.com/docs/developer-guide/

Project Layout

hermes-agent/
├── run_agent.py          # AIAgent — core conversation loop
├── model_tools.py        # Tool discovery and dispatch
├── toolsets.py           # Toolset definitions
├── cli.py                # Interactive CLI (HermesCLI)
├── hermes_state.py       # SQLite session store
├── agent/                # Prompt builder, compression, display, adapters
├── hermes_cli/           # CLI subcommands, config, setup, commands
│   ├── commands.py       # Slash command registry (CommandDef)
│   ├── config.py         # DEFAULT_CONFIG, env var definitions
│   └── main.py           # CLI entry point and argparse
├── tools/                # One file per tool
│   └── registry.py       # Central tool registry
├── gateway/              # Messaging gateway
│   └── platforms/        # Platform adapters (telegram, discord, etc.)
├── cron/                 # Job scheduler
├── tests/                # ~3000 pytest tests
└── website/              # Docusaurus docs site

Config: ~/.hermes/config.yaml (settings), ~/.hermes/.env (API keys).

Adding a Tool (3 files)

1. Create tools/your_tool.py:

import json, os
from tools.registry import registry

def check_requirements() -> bool:
    return bool(os.getenv("EXAMPLE_API_KEY"))

def example_tool(param: str, task_id: str = None) -> str:
    return json.dumps({"success": True, "data": "..."})

registry.register(
    name="example_tool",
    toolset="example",
    schema={"name": "example_tool", "description": "...", "parameters": {...}},
    handler=lambda args, **kw: example_tool(
        param=args.get("param", ""), task_id=kw.get("task_id")),
    check_fn=check_requirements,
    requires_env=["EXAMPLE_API_KEY"],
)

2. Add import in model_tools.py → _discover_tools() list.

3. Add to toolsets.py → _HERMES_CORE_TOOLS list.

All handlers must return JSON strings. Use get_hermes_home() for paths, never hardcode ~/.hermes.

Adding a Slash Command

1. Add CommandDef to COMMAND_REGISTRY in hermes_cli/commands.py
2. Add handler in cli.py → process_command()
3. (Optional) Add gateway handler in gateway/run.py

All consumers (help text, autocomplete, Telegram menu, Slack mapping) derive from the central registry automatically.

Agent Loop (High Level)

run_conversation():
  1. Build system prompt
  2. Loop while iterations < max:
     a. Call LLM (OpenAI-format messages + tool schemas)
     b. If tool_calls → dispatch each via handle_function_call() → append results → continue
     c. If text response → return
  3. Context compression triggers automatically near token limit

Testing

source venv/bin/activate  # or .venv/bin/activate
python -m pytest tests/ -o 'addopts=' -q   # Full suite
python -m pytest tests/tools/ -q            # Specific area

• Tests auto-redirect HERMES_HOME to temp dirs — never touch real ~/.hermes/
• Run full suite before pushing any change
• Use -o 'addopts=' to clear any baked-in pytest flags

Commit Conventions

type: concise subject line

Optional body.

Types: fix:, feat:, refactor:, docs:, chore:

Key Rules

• Never break prompt caching — don't change context, tools, or system prompt mid-conversation
• Message role alternation — never two assistant or two user messages in a row
• Use get_hermes_home() from hermes_constants for all paths (profile-safe)
• Config values go in config.yaml, secrets go in .env
• New tools need a check_fn so they only appear when requirements are met