orkas-ai/orkas/88aca13869d9

Trust tier 2
  • community
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

Install

npx vskill@latest install orkas-ai/orkas/88aca13869d9

SKILL.md

GitHub

Use gh as the primary interface for GitHub data and actions. Keep this skill focused on basic operations and command safety, not maintainer triage strategy.

Before Running Commands

  • Confirm gh exists and is authenticated when live GitHub data is required.
  • Prefer --repo owner/repo unless the current directory is clearly inside the intended GitHub repository.
  • Use --json and --jq when the result will be summarized or filtered.
  • Read-only inspection never needs approval. Resolve the repository, exact object, current state, policy, and relational targets before a write.
  • A current user request authorizes the exact write, direct target, relational target, and stated conditions it specifies. Do not add a second confirmation after preflight succeeds.
  • If the repository or target is unresolved, ask only for the missing target and retain the action authority already granted by the request.
  • When owner/repository itself is missing, ask directly for owner/repo; do not substitute a local checkout path, account hint, account switch, or organization search for that canonical target.
  • Bound target discovery to supplied references and the current repository remote. If those do not resolve owner/repository, ask immediately; do not enumerate accounts, organizations, or unrelated repositories.
  • Ask for new approval only when execution requires a material scope expansion, such as a different repository or object, an extra write, force mode, or admin override. For force/admin/policy-bypass decisions, leave the choice unselected or default to stopping with state unchanged; do not offer a speculative retry or preselect escalation.
  • Honor authentication, permission, deletion, billing, or other platform-required confirmation gates exactly once; do not duplicate them in prose.
  • For a finite maintainer request, batch independent preflight reads in one tool round, authorized independent writes in the next, and final state verification in one last read round. Do not interleave narrative plan updates or repeat unchanged reads.
  • Choose exactly one remote branch-deletion mechanism. If gh pr merge --delete-branch is used, verify absence afterward instead of issuing another delete; otherwise merge without that flag and delete the resolved head ref once.

Common Commands

bash
gh auth status
gh repo view --json nameWithOwner,url,defaultBranchRef
gh issue list --repo owner/repo --state open --limit 50 --json number,title,author,labels,updatedAt,url
gh issue view 123 --repo owner/repo --json number,title,author,body,comments,labels,state,url
gh pr list --repo owner/repo --state open --limit 50 --json number,title,author,isDraft,reviewDecision,mergeStateStatus,url
gh pr view 55 --repo owner/repo --json number,title,state,author,body,comments,files,commits,statusCheckRollup,url
gh pr diff 55 --repo owner/repo --patch
gh pr checks 55 --repo owner/repo
gh run list --repo owner/repo --limit 20
gh run view RUN_ID --repo owner/repo --log-failed
gh release list --repo owner/repo --limit 20
gh api repos/owner/repo/pulls/55 --jq '{title, state, user: .user.login}'
gh api --method DELETE repos/owner/repo/git/refs/heads/feature-branch

For a write not authorized by the current request, show the exact repository, object, command, and effect before asking. For an authorized write, execute once target resolution and preconditions succeed; pause only for a material scope expansion or a platform-required gate. Independent commands may run in parallel, but never parallelize actions whose target or safety depends on another command's result.

Default Output

When reporting GitHub data, include:

  • Repository or URL inspected
  • Commands or data sources used
  • Current state and relevant IDs
  • Any action taken, or the exact command you would run next
  • Permission or authentication blockers, if any

Security

  • Security scanPassed, 96 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
2,163
Repository forks
95
7-day trend score
0

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/orkas-ai/orkas/88aca13869d9/badge)](https://verified-skill.com/skills/orkas-ai/orkas/88aca13869d9)

Something wrong with this skill or its page? Tell us.