add-skill

Capture a useful pattern from the current conversation and contribute it to the diegocanepa/agent-skills repository via a Pull Request.

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "add-skill" with this command: npx skills add diegocanepa/agent-skills/diegocanepa-agent-skills-add-skill

➕ Add Skill from Conversation

Distill a reusable pattern or solution from the current conversation into a new skill and submit it to the shared repository.

When to Use

Use this skill when:

  • You've solved a problem an agent will likely encounter again.
  • The user says "save this as a skill", "add this to our skills", or similar.
  • A new workflow, pattern, or integration approach has been developed.

Workflow

Step 1: Identify the Skill

Ask the user (if not already clear):

  • Name: What should this skill be called? (use kebab-case, e.g. postgres-migrations)
  • Description: One sentence — what does the skill do and when should an agent use it?
  • Category for docs/SKILLS.md: Orchestration / Development / VCS / Utilities?

Step 2: Extract Key Learnings from the Conversation

Review the conversation and extract:

  • The problem that was solved.
  • The approach or solution (tools, patterns, steps).
  • Any caveats, edge cases, or prerequisites.

Present a brief summary to the user before writing the file.

Step 3: Create the Skill using skill-creator

Use the skill-creator skill to scaffold and write the SKILL.md content for skills/<skill-name>/SKILL.md.

The skill-creator will guide the structure. Make sure the generated SKILL.md includes:

  • name and description frontmatter fields (required by the CLI).
  • Self-contained instructions that an agent can follow without prior context.
  • Relative links to other skills if applicable (e.g. [vcs-commit](../vcs-commit/SKILL.md)).

Quality check before proceeding:

Skill Content Review:
- [ ] Frontmatter has `name` and `description`
- [ ] Instructions are self-contained and clear
- [ ] No hardcoded values that should be user-configurable
- [ ] Links to other skills use relative paths
- [ ] An agent could follow this with no prior context

If any check fails → improve the content → re-check.

Step 4: Create the Branch

Use vcs-branch:

feat/add-skill-<skill-name>

Step 5: Commit and Push

Commit using vcs-commit:

git add skills/<skill-name>/SKILL.md
git commit -m "feat(skills): add <skill-name> skill"
git push -u origin feat/add-skill-<skill-name>

Step 6: Update the Skill Catalog

Add the new entry to docs/SKILLS.md under the appropriate category:

| **[<skill-name>](../skills/<skill-name>/SKILL.md)** | <description> |

Commit this as part of the same or a follow-up commit.

Step 7: Create the Pull Request

Use vcs-change-request to open a PR via the GitHub MCP against the skills repository:

  • Owner: diegocanepa
  • Repo: agent-skills
  • URL: https://github.com/diegocanepa/agent-skills
  • Head branch: feat/add-skill-<skill-name>
  • Base branch: main

PR description template:

## Summary

Adds a new skill: **<skill-name>**

<Description of what the skill does and when to use it.>

## Checklist
- [ ] Frontmatter with `name` and `description`
- [ ] Added to `docs/SKILLS.md`
- [ ] Instructions are self-contained

Step 8: MITM — Confirm with User

"The skill <skill-name> is ready and the PR has been submitted. Does the content look correct?"

Feedback Loop: If the USER wants to refine → update the file → amend/push → re-present.


Notes

  • Never push directly to main.
  • The skill file must be self-contained — the entire diegocanepa/agent-skills repo can be installed by anyone, so no project-specific assumptions.
  • The description frontmatter is the most important field — it's what the agent scans to decide whether to use the skill.

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Automation

add-skill

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

skill-improvement

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

mermaid-diagrams

No summary provided by upstream source.

Repository SourceNeeds Review
Automation

vcs-branch

No summary provided by upstream source.

Repository SourceNeeds Review