agentsclimarketplace

Plan archive

Skill arndvs/ctrlshft/skills/plan-archive

Use after merging a PR or during periodic cleanup to archive plan-mode files by linking them to merged PRs.From its SKILL.md

Install
npx -y skills add arndvs/ctrlshft --skill plan-archive

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.7 KB, 683 tokens by cl100k_base, as published. Nobody here has run it

Plan Archive

If running interactively (human present), output "Read Plan Archive skill." to acknowledge. If running with --dangerously-skip-permissions (AFK/unattended), skip acknowledgement and proceed directly.

Archives ~/.claude/plans/*.md files by matching them to merged PRs via file-overlap analysis, then moves them into organized directories with metadata linking back to the PR, Linear tickets, and commits.

When to Use

  • After merging a PR: archive the plans it consumed
  • Periodically: backfill to clean up accumulated plan files
  • Before auditing: see what active plans exist and flag orphans

Modes

Audit (read-only)

List all active plans with last-modified dates. Good for seeing what's accumulated.

python3 skills/plan-archive/plan-archive.py --mode=audit

Archive one PR (dry-run by default)

Match plans to a specific merged PR and archive the bundle:

# Dry-run — see what would move
python3 skills/plan-archive/plan-archive.py --mode=archive-pr --pr=42

# Execute — actually move files
python3 skills/plan-archive/plan-archive.py --mode=archive-pr --pr=42 --execute

Backfill (scan recent merged PRs)

Scan all merged PRs within a window and archive matching plans:

# Dry-run — see what would be archived
python3 skills/plan-archive/plan-archive.py --mode=backfill --since=30d

# Execute
python3 skills/plan-archive/plan-archive.py --mode=backfill --since=2w --execute

Output

Archived plans go to ~/.claude/plans/archive/by-pr/PR-<num>-<branch-slug>/ with a _meta.yaml containing:

  • PR number, URL, title, branch, merge date, merge SHA
  • Linear ticket references (CC-NNN pattern)
  • List of archived plan filenames
  • Cross-references to other PRs that matched the same plans

Options

FlagDescription
--modeaudit, archive-pr, or backfill (required)
--prPR number (required for archive-pr)
--sinceTime window for backfill: 30d, 2w, 12h (default: 30d)
--executeActually move files (default: dry-run)
--repoOverride GitHub repo (default: cwd's repo)
--plans-dirOverride plans directory (default: ~/.claude/plans)
--limitMax PRs to fetch for backfill (default: 200)

How Plan Matching Works

  1. Fetches the PR's diff file list via gh pr diff --name-only
  2. Parses each plan's ## Files section for declared file paths
  3. Normalizes paths to repo-relative form
  4. Matches plans with >= 1 file overlap with the PR diff
  5. Plans are assigned to the earliest-merged PR that claims them (no double-archive)

What ships with it: 1 file

14.3 KB alongside SKILL.md, 1 of them executable

Keep looking

Skills are one crate of 326,764. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.