ghx
Skill194 starsChanged 4 days agoApache-2.0View on GitHub
Guidance for safe, reliable GitHub CLI workflows across issues, pull requests, and reviews.
- Reads credentials
- Commits and pushes
In this repository18 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- 3d-production-workflowSkill
- code-health-auditSkill
- code-reviewSkill
- docxSkill
- email-workflowSkill
- github-issue-solveSkill
- github-pr-fixSkill
- github-reviewSkill
- holon-runtime-opsSkill
- interlinked-markdown-wikiSkill
- learning-coachSkill
- opsSkill
- pdfSkill
- pptxSkill
- video-productionSkill
- xlsxSkill
The file
We split this skill into the rules it gives an agent
11 rules. About 0 are common, 2 are distinctive, and 9 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
9
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 10
Use this skill as shared guidance for raw gh and gh api usage in GitHub issue, pull request, and review workflows.
- Very specific: line 25
When a task needs Holon runtime internals rather than GitHub CLI guidance
- Distinctive: line 38
Never inline large markdown or JSON payloads on the command line.
- Distinctive: line 39
Always pass long bodies through files:
- Very specific: line 44
Use --body-file - only when stdin is clearly available and intentional.
- Very specific: line 65
Prefer gh api --input <file> for complex request bodies.
- Very specific: line 66
When passing many scalar fields, prefer repeated -f key=value flags over hand-built shell strings.
- Very specific: line 131
Use gh api with a JSON payload file so body text and inline comments stay structured and reproducible.
- Very specific: line 135
Prefer explicit repository targeting via --repo owner/repo in automation.
- Very specific: line 136
Prefer machine-readable --json output when subsequent steps need structured data.
- Very specific: line 137
When collecting many records, remember --paginate for REST endpoints.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
GHX Skill
Summary
Use this skill as shared guidance for raw gh and gh api usage in GitHub issue, pull request, and review workflows.
It focuses on safe command patterns, payload handling, and common collection/publish recipes.
This skill is guidance-only. It does not provide scripts, wrappers, or executable entrypoints.
When To Use
- Collecting issue or PR context directly from GitHub with
gh - Publishing PR bodies, issue comments, or review replies safely
- Looking for reusable GitHub CLI patterns without depending on repo-local scripts
- Standardizing GitHub CLI safety rules across multiple skills
Do Not Use
- As a command or executable tool
- When a task needs Holon runtime internals rather than GitHub CLI guidance
- As a replacement for the task-specific logic in
github-review,github-pr-fix, orgithub-issue-solve
Prerequisites
ghCLI authentication is required.GITHUB_TOKENorGH_TOKENmust have the scopes required by the target repository operations.jqis recommended when you need structured filtering or JSON post-processing.
Safety Rules
Payload handling
- Never inline large markdown or JSON payloads on the command line.
- Always pass long bodies through files:
gh issue create --body-file <file>gh issue comment --body-file <file>gh pr create --body-file <file>gh pr edit --body-file <file>
- Use
--body-file -only when stdin is clearly available and intentional.
Good:
cat > /tmp/comment.md <<'EOF'
## Summary
Detailed markdown with quotes, backticks, and newlines.
EOF
gh issue comment 123 --repo owner/repo --body-file /tmp/comment.mdBad:
gh issue comment 123 --repo owner/repo --body "## Summary
Detailed markdown with quotes, backticks, and newlines."Raw JSON payloads
- Prefer
gh api --input <file>for complex request bodies. - When passing many scalar fields, prefer repeated
-f key=valueflags over hand-built shell strings.
Common Recipes
Collect issue context
gh issue view 123 --repo owner/repo --json number,title,body,state,url,author,createdAt,updatedAt,labels
gh api repos/owner/repo/issues/123/comments --paginateCollect PR context
gh pr view 123 --repo owner/repo --json number,title,body,state,url,baseRefName,headRefName,headRefOid,author,createdAt,updatedAt,mergeable,reviews,changedFiles,additions,deletions
gh pr view 123 --repo owner/repo --json files
gh pr diff 123 --repo owner/repo
gh api graphql -f query='
query($owner:String!, $repo:String!, $number:Int!) {
repository(owner:$owner, name:$repo) {
pullRequest(number:$number) {
reviewThreads(first:100) {
nodes {
isResolved
comments(first:100) {
nodes { id body path line author { login } }
}
}
}
}
}
}' -F owner=<owner> -F repo=<repo> -F number=<pr_number>Create or update a PR
gh pr create --repo owner/repo --title "Title" --body-file /tmp/pr-body.md --head feature/x --base main
gh pr edit 123 --repo owner/repo --title "Updated title" --body-file /tmp/pr-body.mdPost a PR-level comment
gh api repos/owner/repo/issues/123/comments -X POST --input /tmp/comment.jsonExample payload:
{
"body": "Comment body"
}Reply to a review comment
gh api repos/owner/repo/pulls/123/comments -X POST \
-F in_reply_to=456789 \
-f body='Thanks, fixed in the latest commit.'Post a review with inline comments
Use gh api with a JSON payload file so body text and inline comments stay structured and reproducible.
Notes
- Prefer explicit repository targeting via
--repo owner/repoin automation. - Prefer machine-readable
--jsonoutput when subsequent steps need structured data. - When collecting many records, remember
--paginatefor REST endpoints.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

