Skill
Skill alessiomarcone/agent-fieldbook/plugins/knowledge-pack/skills/skill
Source-backed skills that turn official Claude and Codex training into reliable daily workflows.
npx -y skills add alessiomarcone/agent-fieldbook --skill skillAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 17 days oldThe repository was created 17 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 3 stars3 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Turns a repeated workflow into one narrow, reusable, validated agent skill with reliable triggers and progressive disclosure. Use when the user asks to create, revise, split, review, or test a SKILL.md workflow for Claude, Codex, or another Agent Skills-compatible surface; not when the request is to distribute a bundle as a plugin.
SKILL.md
2.2 KB, as published. Nobody here has run it
Skill
Create a skill around one repeatable user intent and one output contract.
References
Resolve the target product, then read its branch:
- Claude:
references/claude-code-estensioni.md. - Codex:
references/codex-skills.md. - Read
references/ai-fluency-4d-corso.mdwhen delegation, discernment, or diligence rules shape the workflow.
Procedure
- Confirm the workflow is repeatable and belongs in a skill rather than always-on instructions, a subagent, hook, plugin, or MCP integration.
- Define at least three representative triggers and clear non-triggers.
- Specify inputs, one primary output, failure behavior, side-effect policy, and success checks.
- Keep essential procedure in
SKILL.md; place detailed knowledge in directly linkedreferences/; use scripts only for deterministic repeated operations. - Match freedom to risk: guidance for flexible work, templates for repeatable work, scripts for fragile deterministic work.
- Follow the selected product's current naming and packaging rules.
- Validate structure and forward-test triggers, non-triggers, normal cases, boundaries, and failures.
Output contract
Return:
- Decision — why a skill is the right mechanism.
- Contract — triggers, non-triggers, inputs, output, failure behavior, and authority.
- File tree — only required files.
- Implementation — complete files when requested.
- Validation — exact checks and forward-test prompts.
Boundaries
- Do not create an “everything” skill.
- Do not package or publish a bundle of existing workflows; route distribution work to
plugin. - Do not duplicate detailed material in both
SKILL.mdand references. - Do not add repository documentation inside a skill folder.
- Do not claim success before validation and trigger testing.
- Answer in the user's language.