Sync Repo Status

VerifiedCaution

Regenerate REPO_STATUS.md — full snapshot of open PRs, local branches, issues, discussions, and prioritized action queue.

Sby Skills Guide Bot
DevelopmentIntermediate
7/22/2026
Claude Code
#github#pull-requests#issue-triage#repo-status#gh-cli

Recommended for

Automated summary

Regenerates a REPO_STATUS.md file that snapshots a repository's full current state: open pull requests grouped by whose move it is, local branches without PRs, issues grouped by theme, and open discussions, ending with a prioritized action queue.

Strengths

  • Gives a complete point-in-time state rather than an activity delta, which is useful when catching up after time away
  • Sorts PRs by who is blocking (maintainer, contributor, local author) using reviews, comments and unresolved threads
  • Surfaces local branches without PRs, including unpushed work and branches checked out in other worktrees
  • Flags unanswered external issues and likely duplicates so triage effort goes to the right place

Limitations

  • The sample GraphQL queries hardcode the gittower/git-flow-next repository and must be adapted to your own
  • Relies on an authenticated gh CLI plus GitHub MCP tools for review state and discussion threads
  • Strictly read-only — it never posts, labels or replies — and it overwrites REPO_STATUS.md on every run
When to use it

Use it when you need an up-to-date overview of everything open on a repository before a review or triage session.

When not to use it

Don't use it to track recent activity over time or to actually act on GitHub (reply, label, close), which the skill deliberately does not do.

Security analysis

Caution
Quality score80/100

The skill is explicitly read-only against GitHub and only writes a single gitignored snapshot file. Execution risk comes from legitimate but powerful tooling: bash-driven git/gh commands and GitHub API reads. No secrets are exfiltrated, no obfuscated payloads, no destructive filesystem operations, and no safety-disabling instructions. Rated caution rather than safe because of network/CLI usage and the overwrite behavior, but nothing here is malicious.

Findings
  • •Uses Bash with git and the gh CLI, including network access to the GitHub API (gh api graphql, gh pr/issue list)
  • •Overwrites REPO_STATUS.md in the project root each run (via Write); scoped to a single gitignored file, but it is a destructive write with no backup
  • •Runs `git fetch --all --prune`, which deletes stale remote-tracking refs locally
  • •Hardcodes the repository owner/name (gittower/git-flow-next) in GraphQL queries rather than deriving it from the working directory
How this analysis is made

Examples

Refresh the status snapshot
/sync-repo-status
Catch up before a triage session
Run /sync-repo-status so I can see every open PR, issue and discussion, and which ones are waiting on me.
Audit stale branches and PRs
Use /sync-repo-status to regenerate REPO_STATUS.md and tell me which local branches without PRs and which contributor PRs are stale enough to close.

name: sync-repo-status description: Regenerate REPO_STATUS.md — full snapshot of open PRs, local branches, issues, and discussions with merge-readiness and triage assessment allowed-tools: Bash, Read, Write, Glob, mcp__github__list_issues, mcp__github__list_pull_requests, mcp__github__get_pull_request_reviews, mcp__github__get_pull_request_comments

Sync Repo Status

Regenerate REPO_STATUS.md in the project root: a complete point-in-time snapshot of everything open — PRs (grouped by whose move it is), local branches without PRs, issues (grouped by theme, with reply/duplicate/ linked-PR status), and open discussion questions — ending in a prioritized action queue.

Read-only against GitHub: never posts, labels, or replies. Unlike /scan-repo (a windowed activity delta written to .ai/scans/), this produces the full current state, overwriting REPO_STATUS.md each run.

Arguments

/sync-repo-status

Instructions

1. Gather State

Run in parallel where possible:

Local branches (fetch first so ahead/behind counts are accurate):

git fetch --all --prune
git branch -vv --sort=-committerdate

For each local branch except main:

git rev-list --left-right --count main...<branch>   # behind/ahead
git log -1 --format='%cs %s' <branch>               # last commit date + subject

Note branches with no upstream (unpushed work) and branches checked out in other worktrees (marked + in git branch output).

Open PRs:

gh pr list --state open --limit 50 --json number,title,author,headRefName,baseRefName,isDraft,createdAt,updatedAt,reviewDecision,mergeable,statusCheckRollup

PR ↔ issue links (closingIssuesReferences is GraphQL-only — not available via gh pr view --json):

gh api graphql -f query='{ repository(owner: "gittower", name: "git-flow-next") {
  pullRequests(states: OPEN, first: 50) { nodes { number
    closingIssuesReferences(first: 5) { nodes { number } } } } } }'

Per-PR review state — for each open PR fetch reviews, comments, and unresolved review threads:

gh pr view <n> --json author,reviews,comments
gh api graphql -f query='query($pr: Int!) { repository(owner: "gittower", name: "git-flow-next") {
  pullRequest(number: $pr) { reviewThreads(first: 50) { nodes { isResolved
    comments(first: 1) { nodes { author { login } body } } } } } } }' -F pr=<n>

Determine whose move it is: read the last few PR comments and the latest review verdict. Copilot review threads are often left unresolved even after being addressed — check the author's "addressed review" comments before treating unresolved threads as open work.

Open issues:

gh issue list --state open --limit 100 --json number,title,labels,createdAt,updatedAt,author,comments

For unanswered external issues (no maintainer among commenters), read the body (gh issue view <n> --json title,body) to classify and spot duplicates.

Discussions (GraphQL):

gh api graphql -f query='{ repository(owner: "gittower", name: "git-flow-next") {
  discussions(first: 50, states: OPEN) { nodes { number title createdAt
    isAnswered category { name } author { login }
    comments(first: 10) { totalCount nodes { author { login } } } } } }'

Treat comments by repo collaborators and bots as "ours"; everyone else is external.

2. Assess

PRs — sort into four groups:

  • Own PRs, merge candidates: review feedback addressed, mergeable; note behind-main count and anything to verify before merging (e.g. missing docs for new config keys — cross-check docs/gitflow-config.5.md)
  • Contributor PRs waiting on maintainer: no review yet, or resolved and awaiting approval; flag overlaps with own PRs or local branches
  • Contributor PRs waiting on them: changes requested with no author activity since — note how long, suggest nudge/close when stale (>2 months)
  • Local branches without PRs: assess by ahead/behind and age whether to revive, restart from main, or delete

Issues — group by theme (recurring clusters: merge-state/finish bugs, Windows distribution, config/init bugs, AVH parity & feature requests, own backlog/roadmap). For each issue note:

  • replied or unanswered (bold the unanswered ones)
  • duplicate or same-family of another issue
  • linked to an open PR (from closingIssuesReferences) — likely resolved by a pending merge
  • waiting on reporter → nudge/close candidate

Discussions — flag unanswered Q&A threads, replied-but-unmarked answers, and idea threads that connect to open issues or branches.

3. Write REPO_STATUS.md

Overwrite REPO_STATUS.md in the project root. Structure:

# Repo Status Overview

_Snapshot of open PRs, branches, issues, and discussions — generated <date>._

## Open Pull Requests (<n>)
### Own PRs — merge candidates          (table: PR, title, branch state, status)
### Contributor PRs — waiting on maintainer
### Contributor PRs — waiting on them (stale)

## Local branches without PRs           (table: branch, state, assessment)

## Open Issues (<n>), grouped           (one subsection per theme)

## Discussions

## TLDR action queue                    (numbered, most valuable first)

Every PR/issue reference as a markdown link. Keep assessments to one line each — what it is, where it stands, what unblocks it.

REPO_STATUS.md is listed in .gitignore — verify it still is, and never commit the generated file.

4. Report to User

Summarize in the conversation: counts per group, what changed since the previous snapshot if one existed (newly ready PRs, new unanswered issues, items that dropped off), and the top 3 items from the action queue.

From the file to a first result

The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.

View the source and supporting files

This download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.

1. Download the file

Download SKILL.md

From the root of a test project, save the file as .skill-downloads/sync-repo-status-mrwak975/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.

Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.

mkdir -p '.skill-downloads/sync-repo-status-mrwak975' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/sync-repo-status-mrwak975/download' --output '.skill-downloads/sync-repo-status-mrwak975/SKILL.md.part' && mv '.skill-downloads/sync-repo-status-mrwak975/SKILL.md.part' '.skill-downloads/sync-repo-status-mrwak975/SKILL.md'

2. Check prerequisites

Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.

Tools declared by the author (check availability in your agent): Bash, Read, Write, Glob, mcp__github__list_issues, mcp__github__list_pull_requests, mcp__github__get_pull_request_reviews, mcp__github__get_pull_request_comments

No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.

The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.

3. Load the skill

After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.

Claude Code
test -f '.skill-downloads/sync-repo-status-mrwak975/SKILL.md' && test -s '.skill-downloads/sync-repo-status-mrwak975/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/sync-repo-status' && cp -R '.skill-downloads/sync-repo-status-mrwak975' '.claude/skills/sync-repo-status'

OpenSkills

OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.

Official OpenSkills installation
openskills install './.skill-downloads/sync-repo-status-mrwak975'

openskills read 'sync-repo-status-mrwak975'

Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.

For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.

openskills sync

Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.

4. Try a first task

Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:

Read .skill-downloads/sync-repo-status-mrwak975/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.

Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.

Related skills