Turn SOPs, URLs, or stated intent into an installed skill package. Use when the user asks to generate a skill from a document, link, or workflow. Follow SKILL.md, reference/sop-structure-pipeline.md, and reference/generator-worker-spec.md; finish by promoting the draft into the runtime skills directory (see reference/operator-playbook.md) in the same turn.
63
76%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Medium
Suggest reviewing before use
Fix and improve this skill with Tessl
tessl review fix ./jiuwenclaw/jiuwenclaw/resources/agent/jiuwenclaw_workspace/skills/skill-gen-4-enterprise-doc/SKILL.mdInstructions here, reference/, and scripts/ (package scripts/skill_gen/). You obtain SOP plain text, run full structured extraction exactly as in reference/sop-structure-pipeline.md, author the package under get_agent_workspace_dir() / "skills-draft" (per reference/generator-worker-spec.md), then immediately promote it so the skill is installed and loadable (see reference/operator-playbook.md canonical flow).
python3 <ABS>/scripts/skill_generator_cli.py sop-text --sop-file <abs-path> [--out-text <path>] or read the file in-tool; optional --print-raw-chars for length.url-fetch --url '…' [--out-json <path>] then use page text as the SOP body for extraction.parse_sop_raw_text (no skipping structured extraction).skill_gen.sop_parser.parse_sop_file or parse_sop_raw_text with invoke_llm_json (single path; sop-structure-pipeline.md).get_agent_workspace_dir() / "skills-draft" / <skill_name> (when the host exposes that helper) or the equivalent agent-workspace path from the system prompt: SKILL.md (YAML frontmatter first) plus optional reference/, per generator-worker-spec.md. Create skills-draft next to the runtime skills/ folder if it does not exist.skills.import_local with path = absolute path to that draft directory (folder containing SKILL.md). Use force: true if get_agent_skills_dir() / <skill_name> already exists and should be replaced. Do not stop after step 3 and ask the user to import manually; promotion is part of this skill’s default completion.After step 4, the new skill lives under get_agent_skills_dir() / <skill_name> (same directory tree the runtime loads user-installed skills from) and is available like any other installed skill.
Do not silently shrink generator-worker-spec when writing SKILL.md.
skill_generator_cli.py safelyscripts/skill_generator_cli.py.httpx / beautifulsoup4 for url-fetch. Rich file formats need openjiuwen AutoFileParser when installed; otherwise use .md / .txt SOPs.Path tables and skills.import_local: reference/operator-playbook.md.
SKILL.md format (YAML frontmatter first)The generated skill’s SKILL.md must start with --- YAML frontmatter, then ---, then the body. Optional: scripts/skill_gen/validator.py on the draft path before import.
| File | When |
|---|---|
| reference/sop-structure-pipeline.md | Always: SOPStructure extraction. |
| reference/generator-worker-spec.md | Always: how to write a high-quality generated skill — Writing a high-quality SKILL.md (required structure) plus Skill writing guide. |
| reference/operator-playbook.md | CLI, paths, skills.import_local, canonical flow, directory table, SKILL.md frontmatter rules. |
Do not modify this meta-skill’s SKILL.md, reference/sop-structure-pipeline.md, reference/generator-worker-spec.md, reference/operator-playbook.md, or other bundled reference/*.md while generating a user skill. Edit only the target skill under the workspace-relative skills-draft/<name>/ staging tree or the installed skills/<name>/ tree (resolve with get_agent_workspace_dir() / get_agent_skills_dir() when available).
skill_name (kebab-case) from title or URL when the user did not specify.4fd5cfc
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.