watcher-hermes/hermes-skills/agent-bundle

Claude Code and agent skill by watcher-hermes/hermes-skills

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Produce a portable SKILL.md + AGENTS.md + MCP-server blueprint for a workflow, loadable across Claude Code, Cursor, Codex, and compatible agents.

Install

npx vskill@latest install watcher-hermes/hermes-skills/agent-bundle

SKILL.md

The same SKILL.md was published earlier as watcher-hermes/hermes-skills/agent-bundle.

Given a workflow description, produce an agent bundle.

Produce:

  1. SKILL.md. YAML frontmatter with name and description, markdown body with numbered steps. Include progressive-disclosure subresource references if the body is long.
  2. AGENTS.md entry. A few lines to add to the repo's AGENTS.md reflecting any conventions the skill depends on (linter commands, test commands).
  3. MCP server blueprint. Which tools the skill calls via MCP; name, description (Use-when pattern), and input schema.
  4. Cross-agent translations. SkillKit-style notes on how this SKILL.md maps to Cursor rules, Codex .codex.md, Windsurf rules.
  5. Loading path. Where agents will discover this bundle: ~/.anthropic/skills/, ./skills/, ~/.claude/skills/.

Hard rejects:

  • Any SKILL.md whose name is not kebab-case. Breaks discovery.
  • Any SKILL.md without description in frontmatter. Agent runtimes skip it.
  • Any bundle whose MCP tools are not named per Phase 13 · 05 rules.

Refusal rules:

  • If the workflow is a single one-shot prompt, refuse to produce a skill; recommend inline prompt-engineering.
  • If the workflow requires OAuth (e.g. Slack post), flag that the MCP server's first-run elicitation must handle it.
  • If the target agents do not support SKILL.md (some IDEs), recommend translation via SkillKit or similar.

Output: a one-page bundle with the three files sketched, the cross-agent translation notes, and the loading path. End with the single agent to test the bundle in first.

Security

  • Security scanPassed, 98 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
5

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/watcher-hermes/hermes-skills/agent-bundle/badge)](https://verified-skill.com/skills/watcher-hermes/hermes-skills/agent-bundle)

Something wrong with this skill or its page? Tell us.