hermes-plugins
by 42-evey
Custom plugins for hermes-agent — goal management, inter-agent bridge, model selection, cost control
544stars
60forks
Python
Added 5/13/2026
From $49 · 30 days · terms & refunds
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
Maintainer of this skill?
Automated security scan — not a certification.
Installation
# Add to your Claude Code skills
git clone https://github.com/42-evey/hermes-pluginsGetting Started
Guides for using ai agents skills like hermes-plugins.
- Caveman: Cut Claude Token Use by 65%How agent-side prompt compression works, when to use it, and when not to.
- How to Test an AI Skill: A Practical Guide to EvalsEvaluate skills with realistic fixtures, a baseline, and repeatable checks.
- Getting Started with AI SkillsFirst-time install walkthrough for Claude Code, Codex CLI, and ChatGPT.
Security ReportVerified
Last scanned: 5/30/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-05-30T15:59:14.485Z",
"npmAuditRan": true,
"pipAuditRan": true
}Compare with similar skills
See how hermes-plugins compares with popular alternatives.
hermes-plugins
Support Evey — runs 24/7 on free models, hardware costs ~$69/mo: Donate (BTC, ETH, SOL, XRP, DOGE)
Plugins and skills for Hermes Agent. Checked on
Hermes v0.22.5: every plugin passes hermes plugins doctor --ci and hermes plugins validate,
and the skills load on a fresh install running a local model.
Install
A plugin installs one folder at a time and stays off until enabled:
hermes plugins install 42-evey/hermes-plugins/evey-goals --enable
A running gateway keeps the plugin code it loaded at start. Run hermes gateway stop and the
next session starts a fresh one (or hermes gateway restart when the gateway is installed as a
service). hermes plugins list shows what is enabled.
Skills install the same way, by path:
hermes skills install 42-evey/hermes-plugins/skills/software-development/verify-before-done
Or use the whole folder from a clone:
git clone https://github.com/42-evey/hermes-plugins ~/src/hermes-plugins
hermes config set skills.external_dirs "[~/src/hermes-plugins/skills]"
Skills
Workflow skills that fill gaps in Hermes's bundled set. Each links the bundled skills it works
with (systematic-debugging, test-driven-development, requesting-code-review, spike).
| Skill | Use it |
|---|---|
| verify-before-done | Before claiming work is fixed, done, passing or deployed |
| plan-before-building | Before a multi-step build: decide the design, write the plan, get one approval |
| review-feedback | On receiving code review or PR comments, before changing anything |
| delegation-brief | Writing a delegate_task brief, and checking what comes back |
| work-ledger | Work that spans sessions and must survive compaction |
| safe-cleanup | Before deleting, overwriting or killing anything, or a test run on real data |
Plugins
Plugins that need a model call it through ctx.llm, so they run on whatever provider and model
you have configured, local ones included. A plugin that talks to an outside service reads its
address and keys from the environment variables in the last column; while they are unset its
tools say which variable to set. Plugins that use the time of day read Hermes's own timezone
setting (HERMES_TIMEZONE, UTC when unset).
Autonomy and decisions
| Plugin | What it does | Needs |
|---|---|---|
| evey-autonomy | Priority queue, planning and reflection over goals, projects and inbox (autonomous_decide, autonomous_plan, autonomous_reflect) |
optional $HERMES_HOME/evey-autonomy.json, HERMES_AUTONOMY_* |
| evey-council | Three roles (pragmatist, skeptic, contrarian) answer a hard question; a judge synthesizes | — |
| evey-delegate-model | Delegation with an optional per-task model and parallel fan-out | a model grant to pin models (see its docstring) |
| evey-delegation-score | Scores delegation results per model over time | — |
| evey-goals | Read, add, complete and evolve standing goals | — |
| evey-proactive | Surfaces insights to the user during work hours, within an interruption budget | WORK_HOURS (default 9-21) |
| evey-scheduler | The user's events and reminders, in a JSON file | — |
Quality and safety
| Plugin | What it does | Needs |
|---|---|---|
| evey-reflect | Critiques an output and fixes errors before it is sent | — |
| evey-validate | Flags hallucinations and errors in delegated output | — |
| evey-verification | Checks URLs, repos, endpoints and DNS before the agent acts on them | — |
| evey-email-guard | Screens incoming email for prompt injection (patterns plus a model check); unscreened mail is reported suspicious | — |
| evey-sandbox | Reads and searches files in operator-approved folders, with PII scrubbed (no code execution) | $HERMES_HOME/config/sandbox.yaml, SANDBOX_BASE_DIR |
| evey-session-guard | Checkpoints critical context before compression or long tasks | — |
Memory and learning
| Plugin | What it does | Needs |
|---|---|---|
| evey-learner | Extracts lessons from interactions and applies them later | — |
| evey-memory-adaptive | Importance scoring and decay for memories | — |
| evey-memory-consolidate | Daily: extracts facts from Langfuse traces into MEMORY.md and Qdrant | LANGFUSE_HOST, LANGFUSE_PUBLIC_KEY, LANGFUSE_SECRET_KEY, QDRANT_URL, OLLAMA_URL |
| evey-cache | Caches delegation results (24 h TTL, LRU) | — |
| evey-identity | Updates SOUL.md from the agent's experience | — |
| evey-habits | Learns the user's message times, topics and work rhythm | — |
| evey-rag | Vector search over the agent's knowledge base | QDRANT_URL, EMBED_BASE_URL, OPENAI_API_KEY for the embeddings server |
Observability
| Plugin | What it does | Needs |
|---|---|---|
| evey-telemetry | JSON event log of tool calls, delegations and errors, with telemetry_query |
— |
| evey-status | One call for bridge, MQTT, costs, cron and goals | DASHBOARD_URL |
| evey-cost-guard | Budgets and cost breakdowns from Langfuse | LANGFUSE_HOST, LANGFUSE_PUBLIC_KEY, LANGFUSE_SECRET_KEY |
| evey-watchdog | Alerts the operator through ntfy when the agent goes silent in work hours | NTFY_URL, WORK_HOURS |
| evey-digest | Morning digest of costs, cron health, bridge tasks, goals and alerts | Langfuse variables; NTFY_URL to push it |
| evey-mqtt | MQTT event stream: subscribes to the bridge, events and health topics and buffers messages | MQTT_HOST, MQTT_PORT |
Integrations
| Plugin | What it does | Needs |
|---|---|---|
| evey-bridge | Task and message bridge to a Claude Code session through $HERMES_HOME/claude-bridge |
a Claude Code session reading the bridge |
| evey-github | Stars, issues and PRs for one account, and its PRs upstream | GITHUB_OWNER, GITHUB_UPSTREAM_REPO (default NousResearch/hermes-agent) |
| evey-research | Web search, page extraction (research_extract) and a findings library |
SEARXNG_URL; CRAWL4AI_URL optional |
| evey-news | Trending AI news via SearXNG | SEARXNG_URL |
| evey-telegram-ux | Status cards and formatted results for the Telegram gateway | Hermes's Telegram gateway |
| evey-moltbook | Feed, replies and posts on the Moltbook agent network | MOLTBOOK_API_KEY |
| evey-wallet | Crypto balances with new-deposit detection | BTC_ADDRESS, ETH_ADDRESS, SOL_ADDRESS, XRP_ADDRESS, DOGE_ADDRESS (any subset) |
Checking a change
hermes plugins doctor --ci ./<plugin>
hermes plugins validate ./<plugin>
See CONTRIBUTING.md for the rules plugins and skills follow. A full self-hosted stack with these plugins preinstalled: 42-evey/evey-setup.
License
MIT. verify-before-done, plan-before-building, review-feedback and delegation-brief adapt ideas from obra/superpowers (MIT).
Frequently Asked Questions
What is hermes-plugins?
hermes-plugins is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by 42-evey. Custom plugins for hermes-agent — goal management, inter-agent bridge, model selection, cost control. It has 544 GitHub stars.
Is hermes-plugins safe to use?
Yes. hermes-plugins passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
How do I install hermes-plugins?
Clone the repository with "git clone https://github.com/42-evey/hermes-plugins" and add it to your Claude Code skills directory (see the Installation section above).
What programming language is hermes-plugins written in?
hermes-plugins is primarily written in Python. It is open-source under 42-evey on GitHub, so you can review or fork the full source.
Are there alternatives to hermes-plugins?
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh hermes-plugins against similar tools.
Comments (0)
to leave a comment.
No comments yet. Be the first to share your thoughts!
Related Skills
Reverse engineer anything with agents, from app behavior down to native binaries.
445,259
109,916
TypeScript
AI Agentsagent-skillsai
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
299,887
44,298
JavaScript
AI Agentsai-agentsanthropic
Skills for Real Engineers. Straight from my .agents directory.
250,518
21,171
Shell
AI Agents
An agentic skills framework & software development methodology that works.
234,966
20,863
Shell
AI Agentsaibrainstorming