hermes-official-plugins
☤ The official plugins for Hermes Agent, built and maintained by Nous Research: Home Assistant, Spotify, Claude Subscription DirectSDK and more, one install away from the Hermes plugin catalog.
翻訳待ち
リポジトリ情報はまだ収集していません。
No updates detected yet. GitHub releases will appear here.
D.O.E. framework for reliable AI coding agents (Claude Code / Cursor / Codex): CLAUDE.md + directives + secrets linter. Free lite version.
翻訳待ち
Loading the full README.
Read the full README ↗The D.O.E. framework (Directive / Orchestration / Execution) for running AI coding agents — Claude Code, Cursor, Codex, Gemini CLI — reliably on real work.
The problem: an LLM is probabilistic; your business logic isn't. If the agent executes every step itself, errors compound — 90 % per step is 59 % over five steps. D.O.E. moves the deterministic work into scripts and leaves the agent one job: deciding what to run next. Then it makes the system self-annealing: every failure becomes a test, a rule, and a lint.
The plugin lives in plugins/agent-ops-kit-lite/; the MCP server in mcp/ is separate.
| File | What it is |
|---|---|
plugins/agent-ops-kit-lite/templates/CLAUDE.md | The agent's operating instructions: the 3 layers, 6 operating principles, the self-annealing loop, mandatory directive checks. Cursor: paste into .cursorrules; Codex/Gemini: AGENTS.md / GEMINI.md. |
plugins/agent-ops-kit-lite/templates/directives/core/operating-principles.md | The non-negotiables, written to sync unchanged into every repo you run |
plugins/agent-ops-kit-lite/templates/directives/00_template.md | The directive layout (Goal · Inputs · Execution tools · Outputs · Edge cases · Verification) |
plugins/agent-ops-kit-lite/templates/executions/lint_secrets.py | Catches hard-coded API keys/tokens before they reach git (Stripe, GitHub, AWS, Slack, OpenAI, Anthropic, private keys). Zero dependencies. |
python plugins/agent-ops-kit-lite/templates/executions/lint_secrets.py --root . # exit 1 on any finding
/plugin marketplace add skibuff22/agent-ops-kit-lite
/plugin install agent-ops-kit-lite@calyx-ai
(Claude Code 2.1.275+: /plugin install agent-ops-kit-lite --marketplace skibuff22/agent-ops-kit-lite)
Then in any repo:
/agent-ops-kit-lite:aok-init - scaffolds CLAUDE.md, directives/ and the secrets linter (never overwrites existing files)/agent-ops-kit-lite:aok-directive deploy the API - writes a directive for a job from what's actually in the repo/agent-ops-kit-lite:verify-claims post.md - checks a draft claim by claim against your facts.md before it goes public (PASS/FAIL table; Publish Gate lite)/agent-ops-kit-lite:aok-upgrade - what the paid kit addsmcp/publish-gate/ is a zero-dependency MCP server (Node 18+) with two checks for a draft
before it goes public: check_numbers (every number must be in your facts.md or carry a
[src: URL]) and check_links (dead links, checkout links without utm_source).
git clone https://github.com/skibuff22/agent-ops-kit-lite
claude mcp add publish-gate -- node "$PWD/agent-ops-kit-lite/mcp/publish-gate/server.js"
Any MCP client works the same way (command node, argument the path to server.js).
Details: mcp/publish-gate/README.md.
Also listed in the official MCP registry as io.github.skibuff22/publish-gate (version 0.1.0),
packaged as an MCP Bundle attached to the mcp-v0.1.0 release.
The bundle has not been tested in an MCPB client yet; the clone install above is the tested path.
plugins/agent-ops-kit-lite/templates/CLAUDE.md and plugins/agent-ops-kit-lite/templates/directives/ into your repo root.plugins/agent-ops-kit-lite/templates/directives/00_template.md, named after the job
(deploy_api.md). Reference the script that will do the work.The paid Agent Ops Kit adds everything CLAUDE.md's mandatory-checks table refers to:
→ calyx-ai.net — instant download, free updates, 14-day refund.
Built and used daily to run a small engineering company's CRM, invoicing, inventory and calibration platform with a fleet of agents. Every governance rule in the full kit is a real incident turned into a standard.