ms_skill

This skill creates or refactors Codex skills with minimal diffs, updating frontmatter, SKILL.md, and agents/openai.yaml while keeping lean.
  • Python

42

GitHub Stars

1

Bundled Files

3 weeks ago

Catalog Refreshed

2 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 veilstart where the catalogue uses aiagentskills.

npx veilstart add skill tkersey/dotfiles --skill ms

  • SKILL.md6.0 KB

Overview

This skill creates, updates, and refactors Codex skills in-place with minimal diffs. It focuses on tight trigger descriptions, consistent frontmatter, and optional UI metadata while enforcing validation and proof-of-completion steps.

How this skill works

It inspects the target skill folder and applies the smallest possible changes to satisfy a request: updating frontmatter, editing the procedural guidance, adding reusable assets only when needed, and regenerating or verifying agents/openai.yaml. The workflow always runs the provided quick validation script and records the validation command and result as proof of completion.

When to use it

  • Create a new Codex skill for a specific intent and keep it lean.
  • Tighten or clarify trigger text in a skill's frontmatter based on usage evidence.
  • Refactor a skill to move deep details into a references folder and simplify the main guidance.
  • Regenerate or verify agents/openai.yaml to keep UI metadata consistent.
  • Make minimal in-place edits to an existing skill without adding extra docs or churn.

Best practices

  • Change only what is necessary—preserve unrelated files and keys to minimize diffs.
  • Keep frontmatter minimal: name and description are required; include trigger cues in description.
  • Use the provided generator to create agents/openai.yaml when needed and follow its field constraints.
  • Run the quick validation command before finishing and record the exact command and pass/fail result.
  • Add scripts/references/assets only when they enable real reuse across edits or workflows.

Example use cases

  • Add a new workflow step and update the frontmatter trigger text to reflect supported file types.
  • Refactor a skill's long procedural content into a concise decision tree and place details in references.
  • Regenerate agents/openai.yaml after changing UI fields, then record openai_yaml: regenerated.
  • Verify an unchanged agents/openai.yaml and record openai_yaml: verified unchanged.
  • Create a new skill scaffolded with minimal resources and validate it with quick_validate.py.

FAQ

Record the agents/openai.yaml disposition (regenerated, verified unchanged, or not present) and include the exact quick_validate.py command plus its pass/fail result.

How do you keep diffs minimal?

Identify the smallest set of files and frontmatter keys that must change, avoid formatting churn, and only add resource folders when they provide reuse.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational