# Skill Authoring Coach > Use this skill when the user asks to create, improve, review, package, genericise, or submit an Agent Skill, SKILL.md file, skill metadata, or reusable agent instruction bundle. - Skill: `kody-w/skill-authoring-coach` (Agent Skill, multi-file: 3 files) - Install (CLI): `npx skillmds@latest add kody-w/skill-authoring-coach` - Raw SKILL.md: https://api.skillmd.com/api/skills/kody-w/skill-authoring-coach/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: AI & ML - Author: kody-w (https://skillmd.com/u/kody-w) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/kody-w/skill-authoring-coach --- ## Run this — do not improvise This capability's deterministic implementation is a RAPP single-file agent, linked beside this file as `skill_authoring_coach_agent.py` and embedded as the fenced Python below (sha256 a034e6ed701e5278…; a byte-exact copy is also vaulted in the capsule comment at the end of this file). On a host with sandbox execution, run the linked file directly — if it is missing, write the fence contents verbatim to `skill_authoring_coach_agent.py` first: ```bash python3 skill_authoring_coach_agent.py '{"key": "value"}' # arguments as one JSON object echo '{"key": "value"}' | python3 skill_authoring_coach_agent.py # or on stdin python3 skill_authoring_coach_agent.py --tool # emit the JSON tool contract ``` Treat stdout as a tool result. If it reports missing or unresolved inputs, stop and collect them. If it returns `steps`, execute those steps in order exactly as returned; if it returns `instructions`, follow them with the supplied inputs. Otherwise use the result verbatim. Do not invent behavior beyond that output. On a host without code execution, treat the Parameters schema and the code below as the exact specification and never paraphrase a step. Never edit inside the generated markers; a converter-equipped host can instead restore the original file checksum-verified with the installed `rapp-agent-converter/scripts/toast.py convert SKILL.md --to agent`. ````python # rapp:deterministic """SkillAuthoringCoach -- Use this skill when the user asks to create, improve, review, package, genericise, or submit an Agent Skill, SKILL.md file, skill metadata, or reusable agent instruction bundle. Generated by the rapp skill from skill-authoring-coach. The RCI capsule at the bottom of this file carries the full original; `toast.py convert` restores it byte-exact.""" import json import re import sys try: from agents.basic_agent import BasicAgent except ImportError: # running OUTSIDE a brainstem -- stay executable anyway. class BasicAgent: # noqa: D101 - minimal stand-in, same contract def __init__(self, name=None, metadata=None): if name: self.name = name if metadata: self.metadata = metadata def perform(self, **kwargs): return "Not implemented." def system_context(self): return None def to_tool(self): return {"type": "function", "function": { "name": self.name, "description": self.metadata.get("description", ""), "parameters": self.metadata.get("parameters", {})}} # The procedural layer, verbatim from the source capability. INSTRUCTIONS = '# Skill Authoring Coach\n\nUse this skill to help users design, review, genericise, and package reusable Agent Skills.\n\n## Purpose\n\nA good skill is not a long prompt dump. It is a focused, reusable capability with:\n\n- A clear trigger description.\n- Practical step-by-step instructions.\n- Guardrails and edge cases.\n- Optional references, assets, or scripts.\n- A compact structure that an agent can load when needed.\n\n## Skill vs reference vs template\n\nUse these distinctions:\n\n- **Skill** - Teaches the agent how to perform a repeatable task or workflow.\n- **Reference** - Supplies detailed information the skill may need, such as policies, rubrics, examples, taxonomies, or checklists.\n- **Template** - Supplies a reusable output shape or file structure.\n- **Script** - Performs deterministic work that should not rely on prose instructions alone.\n\nIf the user tries to put everything into `SKILL.md`, recommend moving detailed lookup material into `references/` and static output structures into `assets/`.\n\n## Recommended skill shape\n\n```text\n/\n├── SKILL.md\n├── references/ optional\n├── assets/ optional\n└── scripts/ optional\n```\n\nThe `SKILL.md` should contain:\n\n1. Frontmatter with `name` and `description`.\n2. A short explanation of when to use the skill.\n3. Inputs to look for.\n4. Workflow steps.\n5. Output formats.\n6. Guardrails and quality checks.\n\n## Workflow\n\n1. Identify the workflow the user wants to package.\n2. Decide whether it should be a skill, reference, template, script, or combination.\n3. Write or improve the skill trigger description.\n4. Keep instructions concise and operational.\n5. Move long examples and detailed reference material out of `SKILL.md` where appropriate.\n6. Check for sensitive, organisation-specific, or non-reusable content.\n7. Produce a candidate folder structure and files.\n\n## Review rubric\n\nScore candidate skills from 0 to 5 on:\n\n1. **Trigger clarity** - Will the agent know when to use it?\n2. **Task focus** - Is it one coherent capability?\n3. **Instruction quality** - Are the steps clear and repeatable?\n4. **Portability** - Can others use it without private assumptions?\n5. **Safety and evidence discipline** - Does it avoid invented claims, secret leakage, or risky actions?\n6. **Packaging quality** - Are references, assets, and scripts separated cleanly?\n\n## Output format for review\n\n```markdown\n## Skill review\n\nOverall recommendation: [Submit / Revise / Keep private]\n\n| Dimension | Score /5 | Notes |\n|---|---:|---|\n\n## Required changes\n\n1. [Change]\n2. [Change]\n3. [Change]\n\n## Suggested folder structure\n\n```text\n[folder tree]\n```\n```\n\n\n## References\n\nThis skill includes supporting reference material. Read the relevant reference file when the task needs additional structure, rubric detail, examples, or checklist support.\n\n- `references/skill-design-checklist.md` - use this when additional structure, examples, or checks are useful for the task.\n\n## Quality checklist\n\nBefore responding, check:\n\n- The proposed skill has a clear reusable job.\n- The trigger description is precise.\n- Organisation-specific material is removed or parameterised.\n- Long reference content is not overloaded into the main skill file.\n- The result is practical for a maker to copy into a skills repository.' # Ordered commands lifted verbatim from the capability's own documentation. STEPS = [] class SkillAuthoringCoachAgent(BasicAgent): def __init__(self): self.name = 'SkillAuthoringCoach' self.metadata = { "name": "SkillAuthoringCoach", "description": "Use this skill when the user asks to create, improve, review, package, genericise, or submit an Agent Skill, SKILL.md file, skill metadata, or reusable agent instruction bundle.", "parameters": { "type": "object", "properties": {}, "required": [] } } super().__init__(name=self.name, metadata=self.metadata) def perform(self, **kwargs): # toaster:generated-perform return json.dumps({"status": "ok", "instructions": INSTRUCTIONS, "inputs": kwargs, "note": "Prose-only capability: follow INSTRUCTIONS " "with the given inputs."}, indent=2) if __name__ == "__main__": # echo '{"arg": "value"}' | python3 skill_authoring_coach_agent.py # python3 skill_authoring_coach_agent.py '{"arg": "value"}' # python3 skill_authoring_coach_agent.py --tool # emit the JSON tool contract _a = sys.argv[1:] if _a and _a[0] == "--tool": print(json.dumps(SkillAuthoringCoachAgent().to_tool(), indent=2)) else: _raw = _a[0] if _a else (sys.stdin.read().strip() or "{}") print(SkillAuthoringCoachAgent().perform(**json.loads(_raw))) # rci-capsule:v1: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 ```` # Skill Authoring Coach Use this skill to help users design, review, genericise, and package reusable Agent Skills. ## Purpose A good skill is not a long prompt dump. It is a focused, reusable capability with: - A clear trigger description. - Practical step-by-step instructions. - Guardrails and edge cases. - Optional references, assets, or scripts. - A compact structure that an agent can load when needed. ## Skill vs reference vs template Use these distinctions: - **Skill** - Teaches the agent how to perform a repeatable task or workflow. - **Reference** - Supplies detailed information the skill may need, such as policies, rubrics, examples, taxonomies, or checklists. - **Template** - Supplies a reusable output shape or file structure. - **Script** - Performs deterministic work that should not rely on prose instructions alone. If the user tries to put everything into `SKILL.md`, recommend moving detailed lookup material into `references/` and static output structures into `assets/`. ## Recommended skill shape ```text / ├── SKILL.md ├── references/ optional ├── assets/ optional └── scripts/ optional ``` The `SKILL.md` should contain: 1. Frontmatter with `name` and `description`. 2. A short explanation of when to use the skill. 3. Inputs to look for. 4. Workflow steps. 5. Output formats. 6. Guardrails and quality checks. ## Workflow 1. Identify the workflow the user wants to package. 2. Decide whether it should be a skill, reference, template, script, or combination. 3. Write or improve the skill trigger description. 4. Keep instructions concise and operational. 5. Move long examples and detailed reference material out of `SKILL.md` where appropriate. 6. Check for sensitive, organisation-specific, or non-reusable content. 7. Produce a candidate folder structure and files. ## Review rubric Score candidate skills from 0 to 5 on: 1. **Trigger clarity** - Will the agent know when to use it? 2. **Task focus** - Is it one coherent capability? 3. **Instruction quality** - Are the steps clear and repeatable? 4. **Portability** - Can others use it without private assumptions? 5. **Safety and evidence discipline** - Does it avoid invented claims, secret leakage, or risky actions? 6. **Packaging quality** - Are references, assets, and scripts separated cleanly? ## Output format for review ```markdown ## Skill review Overall recommendation: [Submit / Revise / Keep private] | Dimension | Score /5 | Notes | |---|---:|---| ## Required changes 1. [Change] 2. [Change] 3. [Change] ## Suggested folder structure ```text [folder tree] ``` ``` ## References This skill includes supporting reference material. Read the relevant reference file when the task needs additional structure, rubric detail, examples, or checklist support. - `references/skill-design-checklist.md` - use this when additional structure, examples, or checks are useful for the task. ## Quality checklist Before responding, check: - The proposed skill has a clear reusable job. - The trigger description is precise. - Organisation-specific material is removed or parameterised. - Long reference content is not overloaded into the main skill file. - The result is practical for a maker to copy into a skills repository.