agentleFS
Sign inStart free
AhmedMerzaAhmedMerza/agent-skills/skills/handover-resume/SKILL.md

handover-resume

Skill2 starsChanged 41 days agoMITView on GitHub

Reload a saved handover plan from .claude/handover/, re-sync it against current code, and re-anchor this session to continue the work. Usage: handover-resume [slug] (optional; lists available plans if omitted or ambiguous)

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 5 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If $ARGUMENTS names a slug, read .claude/handover/<slug>.md.

    line 10
  2. Very specific:

    If no argument: run git branch --show-current, derive a slug, and try .claude/handover/<slug>.md.

    line 11
  3. Very specific:

    If neither resolves, list the files in .claude/handover/ (with their # Title and Status: lines) and ask the user which one — do not guess.

    line 12
  4. Distinctive:

    Read the plan fully.

    line 21
  5. Distinctive:

    Check the branch matches; if git status shows relevant uncommitted work, reconcile it against the plan's checkbox state (something may …

    line 24
  6. Distinctive:

    Check off steps that are already done, and update the Status line.

    line 31
  7. Distinctive:

    Leave the doc more accurate than you found it — whether or not you go on to execute it this session.

    line 37
  8. Distinctive:

    If every step is already checked (or the Status line says the work is complete), don't invent new work — report that the plan is done and …

    line 46

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

You are resuming work from a previously saved handover plan (written by /handover-save).

1. Locate the plan

  • If $ARGUMENTS names a slug, read .claude/handover/<slug>.md.
  • If no argument: run git branch --show-current, derive a slug, and try .claude/handover/<slug>.md.
  • If neither resolves, list the files in .claude/handover/ (with their # Title and **Status:** lines) and ask the user which one — do not guess. If the directory is missing or empty, tell the user there are no saved handovers yet.

2. Re-anchor — verify before trusting

The doc reflects what was true when it was saved; the codebase may have moved on. Before you continue:

  • Read the plan fully.
  • Spot-check the Key code anchors — confirm the cited file:line references still point at what the doc claims (Read/grep them). Note any that have drifted.
  • Check the branch matches; if git status shows relevant uncommitted work, reconcile it against the plan's checkbox state (something may already be done that isn't checked off).

Persist the corrections — don't just report them. As you verify, update the doc in place so it stays a trustworthy resume point:

  • Refresh drifted file:line anchors to their current values (grep the symbol, rewrite the number).
  • Check off steps that are already done, and update the Status line.
  • For structural drift — new code that expands or invalidates the plan, not just moved lines (e.g. a sibling feature now needs the same fix) — do NOT silently rewrite the plan. Add a dated note under Open questions (or a short ## Drift since save section) describing what changed and its impact, and flag it in your briefing so the user decides how to adjust scope.

Leave the doc more accurate than you found it — whether or not you go on to execute it this session.

3. Report back, then continue

Give a short briefing:

  • Where we are: the Status line + what's actually done vs pending (corrected for any drift).
  • Any drift you found: stale anchors, already-completed steps, changed assumptions.
  • Next step: the first unchecked, unblocked item.

If every step is already checked (or the Status line says the work is complete), don't invent new work — report that the plan is done and ask the user what they want next.

Otherwise proceed with that next step (or ask which item to start on if several are unblocked). As you complete steps, keep the handover doc's checkboxes updated so it stays an accurate resume point.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

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.

Check your setup, free