- Home
- Skills
- Leegonzales
- Aiskills
- Skill Extractor
skill-extractor_skill
- Python
20
GitHub Stars
4
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill leegonzales/aiskills --skill skill-extractor- CHANGELOG.md364 B
- LICENSE1.1 KB
- README.md2.0 KB
- SKILL.md2.6 KB
Overview
This skill extracts reusable automation skills from repeated conversation patterns. It guides you from example conversations through pattern identification to a ready-to-use skill file and test prompt. The goal is a reliable, repeatable workflow that others can apply without re-teaching the steps.
How this skill works
You provide one or more conversation excerpts or descriptions of a repeated task. The extractor identifies trigger phrases, inputs, transformations, outputs, and quality criteria. It validates the inferred workflow with a confirmation prompt, generates a complete skill file plus a test prompt and usage examples, then recommends iterative tests and refinements.
When to use it
- You keep asking the assistant the same thing multiple times
- You want to convert a manual prompt sequence into a reusable skill
- You have several examples and want a generalized workflow
- You need clear triggers and success criteria for automation
- You want a testable prompt and usage examples for handoff
Best practices
- Provide 3+ high-quality examples to surface consistent patterns
- Include both successful and failed examples to define quality boundaries
- Describe explicit trigger phrases users use to start the task
- State precise input formats and expected output structure
- Run the generated test prompt on fresh inputs and iterate quickly
Example use cases
- Turn meeting note excerpts into a weekly update generator with Accomplishments/In Progress/Blockers sections
- Convert repeated code review requests into an analyzer that lists issues and suggested fixes
- Extract a content brief generator from multiple blog-outline requests into a parameters→draft generator
- Synthesize many customer feedback snippets into a themes-and-priority synthesis skill
- Create a transformation that turns raw research notes into a structured summary and action list
FAQ
Paste conversation excerpts, describe the repeated task with examples, or supply multiple variants; both free text and bulleted examples work.
How do I know a pattern is viable to extract?
A strong candidate appears 3+ times with the same structure, has clear input→output mapping, and produces repeatable quality when prompted.