skills/handoff/SKILL.md
Make sure to use this skill whenever the user types /handoff (with or without a filename argument), says "handoff", "save state", "context rot", "save session", "create a handoff", "I need a clean start", wants to snapshot progress before clearing context or switching sessions, is approaching the context limit (300–400k tokens), or wants to delegate the current session state to another agent. Also invoke for RESUME mode: when the user says "load handoff", "resume from [file]", "continue where we left off", "pick up where I left off", "load the handoff at [path]", references a .claude/handoffs/ file path, or says there is a handoff file from a prior session. Creates or loads a structured JSON snapshot capturing goals, decisions, completed steps, pending work, constraints, and modified files so work can continue cleanly in a new session.
npx skillsauth add psenger/ai-agent-skills handoffInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Saves or loads a structured JSON snapshot of session state. Use the schema in references/schema.md for all fields.
Use the user's argument if provided (e.g. /handoff auth-work.json). Otherwise default
to .claude/handoffs/YYYY-MM-DD-HHMMSS-<slug>.json in the project root, where <slug>
is 2–4 words from the task goal (e.g. auth-refactor, payment-api). Create
.claude/handoffs/ if it doesn't exist.
Review the full conversation and all modified files, then extract:
Write the JSON using the schema in references/schema.md. Do not omit any top-level
field — use [] if there is nothing to report.
Handoff written to: <filename>
Summary:
- Task: <goal>
- Completed: <N> steps
- Pending: <N> steps
- Constraints: <N>
- Open issues: <N>
- Modified files: <N>
Next steps:
1. Run /clear to wipe the context window
2. Start a new session
3. Paste this prompt to resume:
"<resume_prompt value>"
Do not run
/clearyourself — the user must do this manually.
Use the user's argument if given. Otherwise read .claude/handoffs/ and pick the most
recent file by filename timestamp. If the directory is empty or missing, tell the user.
Read the file fully. Then report:
Loaded handoff: <filename>
- Task: <goal>
- Completed: <N> steps
- Pending: <N> steps
- First step: <pending_steps[0]>
- Open issues: <N>
- Constraints: <N>
Start immediately with pending_steps[0]. Keep constraints, discovered_issues, and
decisions visible as you work — don't re-litigate choices already made.
references/schema.md — full JSON schema, field rules, and a complete exampletesting
Exports a single Obsidian vault note and all its linked images into a portable zip or tar.gz archive, preserving vault-root-relative paths so the archive unpacks correctly anywhere. Use only when the user explicitly invokes /export-vault-note.
development
Provides the audit checklists, severity criteria (blocking/warning/suggestion), and artifact patterns needed to properly review Agent OS profiles and standards. Always invoke this skill before auditing - without it you can only give generic feedback, not structured severity-tagged findings. Invoke when the user pastes a standard and asks if it is good or what is wrong with it; when the user asks to review, audit, validate, or critique an agent-os profile or standard; or when the user mentions "agent-os profile", "agent-os standard", or "my agent-os setup" in a review or validation context.
development
Converts transcripts, video summaries, meeting notes, brainstorming sessions, strategy documents, and rough notes into polished Obsidian-flavored Markdown. Activates when creating or editing notes in an Obsidian vault, generating front matter, applying callout blocks, structuring knowledge base articles, or producing developer-facing guides. Also triggers on mentions of Obsidian, front matter, callout blocks, vault organisation, or requests for GitHub-compatible Markdown documents.
documentation
Generate git commit messages, PR titles/descriptions, and changelog entries. Analyzes staged changes, enforces Conventional Commits, scans for sensitive content, links tickets (GitHub Issues / Jira), and updates CHANGELOG.md. Triggers on: "commit", "create a PR", "push", "changelog", "release", or when the user is ready to commit or open a pull request.