managing-changesets

name	managing-changesets
description	Use when working with Changesets — changesets version management covering changeset file inventory, pending version bump analysis, package release readiness, changelog generation status, and monorepo release coordination. Use when auditing pending changesets, investigating release blockers, reviewing version bump strategies, or managing multi-package release workflows.
connection_type	github
preload	false

Changesets Management Skill

Analyze and monitor Changesets version management, pending changes, and release workflows.

MANDATORY: Discovery-First Pattern

Always locate changeset configuration and pending changesets before analyzing release readiness.

Phase 1: Discovery

#!/bin/bash

GH_API="https://api.github.com"

gh_api() {
    curl -s -H "Authorization: Bearer $GITHUB_TOKEN" \
         -H "Accept: application/vnd.github+json" \
         "${GH_API}/${1}"
}

echo "=== Locate Changesets Config ==="
gh_api "orgs/${GITHUB_ORG}/repos?per_page=15&sort=updated" | jq -r '.[].full_name' | while read repo; do
    if gh_api "repos/${repo}/contents/.changeset/config.json" 2>/dev/null | jq -e '.name' >/dev/null 2>&1; then
        echo -e "${repo}\t.changeset/config.json\tFOUND"
    fi
done | column -t | head -15

echo ""
echo "=== Changesets Config ==="
REPO="${GITHUB_ORG}/${GITHUB_REPO}"
gh_api "repos/${REPO}/contents/.changeset/config.json" | jq -r '.content' | base64 -d 2>/dev/null | jq '{
    changelog: .changelog,
    commit: .commit,
    fixed: .fixed,
    linked: .linked,
    access: .access,
    baseBranch: .baseBranch
}'

echo ""
echo "=== Version Packages PR ==="
gh_api "repos/${REPO}/pulls?state=open&per_page=10" | jq -r '
    .[] |
    select(.title | startswith("Version Packages")) |
    "#\(.number)\t\(.title[:50])\t\(.created_at[:10])\t\(.updated_at[:10])"
' | column -t

Phase 2: Analysis

#!/bin/bash

REPO="${1:-${GITHUB_ORG}/${GITHUB_REPO}}"

echo "=== Pending Changesets ==="
gh_api "repos/${REPO}/contents/.changeset" | jq -r '
    .[] |
    select(.name != "config.json" and .name != "README.md" and (.name | endswith(".md"))) |
    "\(.name)\t\(.size) bytes"
' | column -t | head -20

echo ""
echo "=== Changeset Details (latest 5) ==="
gh_api "repos/${REPO}/contents/.changeset" | jq -r '
    [.[] | select(.name != "config.json" and .name != "README.md" and (.name | endswith(".md")))] |
    sort_by(.name) | reverse | .[0:5] | .[].name
' | while read cs; do
    echo "--- ${cs} ---"
    gh_api "repos/${REPO}/contents/.changeset/${cs}" | jq -r '.content' | base64 -d 2>/dev/null | head -10
done | head -30

echo ""
echo "=== Recent Releases (tags) ==="
gh_api "repos/${REPO}/releases?per_page=10" | jq -r '
    .[] |
    "\(.tag_name)\t\(.published_at[:10])\t\(.name[:40])"
' | column -t

echo ""
echo "=== Changeset Summary ==="
PENDING=$(gh_api "repos/${REPO}/contents/.changeset" | jq '[.[] | select(.name != "config.json" and .name != "README.md" and (.name | endswith(".md")))] | length')
VERSION_PR=$(gh_api  | jq )
 
 

Output Rules

• TOKEN EFFICIENCY: Target <=50 lines per output
• Summarize changeset bump types rather than reading all changeset files
• Focus on release readiness and blocking issues

Output Format

Present results as a structured report:

Managing Changesets Report
══════════════════════════
Resources discovered: [count]

Resource       Status    Key Metric    Issues
──────────────────────────────────────────────
[name]         [ok/warn] [value]       [findings]

Summary: [total] resources | [ok] healthy | [warn] warnings | [crit] critical
Action Items: [list of prioritized findings]

Target ≤50 lines of output. Use tables for multi-resource comparisons.

Anti-Hallucination Rules

1. NEVER assume resource names — always discover via CLI/API in Phase 1 before referencing in Phase 2.
2. NEVER fabricate metric names or dimensions — verify against the service documentation or --help output.
3. NEVER mix CLI commands between service versions — confirm which version/API you are targeting.
4. ALWAYS use the discovery → verify → analyze chain — every resource referenced must have been discovered first.
5. ALWAYS handle empty results gracefully — an empty response is valid data, not an error to retry.

Counter-Rationalizations

Shortcut	Counter	Why
"I'll skip discovery and check known resources"	Always run Phase 1 discovery first	Resource names change, new resources appear — assumed names cause errors
"The user only asked for a quick check"	Follow the full discovery → analysis flow	Quick checks miss critical issues; structured analysis catches silent failures
"Default configuration is probably fine"	Audit configuration explicitly	Defaults often leave logging, security, and optimization features disabled
"Metrics aren't needed for this"	Always check relevant metrics when available	API/CLI responses show current state; metrics reveal trends and intermittent issues
"I don't have access to that"	Try the command and report the actual error	Assumed permission failures prevent useful investigation; actual errors are informative

Common Pitfalls

• Missing changesets: PRs without changesets produce no version bump -- enforce via CI check
• Bump types: major/minor/patch in changeset frontmatter controls version increment
• Fixed groups: Fixed packages version together -- one major bump upgrades all
• Linked packages: Linked packages get the same bump type but independent versions
• Version PR conflicts: Version Packages PR can have merge conflicts with new changesets
• Access setting: "restricted" requires npm OTP for publishing -- "public" for public packages
• Pre-release mode: Entering/exiting pre-release mode requires explicit changeset commands
• Monorepo coordination: Changesets track per-package bumps -- verify all affected packages are listed