Agent-Skills.md

Agent Skills: Octocode Skills

Use when working with Agent Skills or SKILL.md folders: find, evaluate, lint, rate, improve, install, or create them across local paths, GitHub, and marketplaces — including description tuning and install targets.

UncategorizedID: bgauryy/octocode-mcp/octocode-skills

Repository

bgauryy/octocode-mcp
bgauryyLicense: MIT
88173

Install this agent skill to your local

pnpm dlx add-skill https://github.com/bgauryy/octocode/tree/HEAD/skills/octocode-skills

Skill Files

Browse the full folder contents for octocode-skills.

Download Skill

Loading file tree…

skills/octocode-skills/SKILL.md

Skill Metadata

Name
octocode-skills
Description
"Use when working with Agent Skills or SKILL.md folders: find, evaluate, lint, rate, improve, install, or create them across local paths, GitHub, and marketplaces — including description tuning and install targets."

Octocode Skills

Evaluate, lint, install, and author Agent Skills: folders with SKILL.md plus optional references/, scripts/, and assets/. Agents load skills progressively, so keep SKILL.md as the compact operating map and route conditional detail out. Flow: UNDERSTAND -> DISCOVER -> INSPECT -> JUDGE -> RECOMMEND -> USER GATE -> ACT -> VERIFY.

Hard rules:

  • Inspect real SKILL.md content before recommending, adapting, installing, or quoting.
  • Identify each candidate by repo/path or absolute path; filter results, never hand over raw search dumps.
  • Gate every install, write, overwrite, symlink, and config change behind user approval.

Stop when one inspected recommendation fits, two High-quality candidates establish a clear top pick, three search angles add nothing new, or a user gate is pending.

Tooling

For Octocode-backed skill research, load references/octocode.md and delegate search mechanics to octocode-research when installed. Keep this skill focused on evaluating, linting, improving, and installing the skill folders found.

Reference Map

Load one reference on demand; each route states when it applies.

  • references/search-playbook.md — when discovering candidates.
  • references/discovery-surfaces.md — when shopping beyond raw GitHub.
  • references/quality-rubric.md — when judging workflow quality.
  • references/quality-signals.md — when ranking by evidence beyond stars.
  • references/output-format.md — when presenting results or deep-dives.
  • references/agent-skills-guide.md — when evaluating, improving, or authoring a skill.
  • references/description-tuning.md — when optimizing a description trigger.
  • references/self-improvement.md — when rating, reviewing, refactoring, or linting a skill.
  • references/skill-lint.md — when linting, updating, or creating a skill (new or existing); documents scripts/skill-lint.mjs.
  • references/hooks.md — when adding, reviewing, or explaining a skill's lifecycle hooks; routes to assets/hooks/ templates.
  • references/install-reference.md — when installing or choosing targets/scopes.
  • references/fetch-and-create-locally.md — when fetching a remote skill into a local folder.
  • references/create-local-skill.md — when creating or synthesizing a local skill.
  • references/recovery.md — when search, fetch, install, or marketplace access fails.

Scripts

  • scripts/skill-lint.mjs — lint skill structure, routing, scripts, and prompt quality; run before reporting any created or edited skill done.
  • assets/hooks/ — copy-paste hook templates; read references/hooks.md before wiring a new hook.

Installation

  • Installing: read references/install-reference.md — inspect scripts/ and hooks before copying third-party skills; confirm provider, scope, copy-vs-symlink, conflicts, and verification plan.
  • Creating: read references/create-local-skill.md — synthesize from evidence, write lean files, include a references/references.md audit trail using references/references-template.md, then run scripts/skill-lint.mjs before reporting done.