agentleFS
Sign inStart free
marylinmarylin/Continuum/skills/recover/SKILL.md

recover

Skill0 starsChanged 7 months agoMITView on GitHub

Recover work from a crashed or interrupted session. Use when Claude crashed mid-task, after an unexpected exit, when you see 'stale sessions detected', or when you lost context and need to figure out what was happening.

We split this skill into the rules it gives an agent

9 rules. 0 are common, 4 are distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 11 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

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

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

    Look for checkpoint files in .lifecycle/checkpoints/

    line 15
  2. Distinctive:

    If found, present cognitive state alongside recovery data — not just "these files changed" but "you were doing X because Y"

    line 16
  3. Very specific:

    Check for wip/ branches: git branch --list 'wip/*'

    line 20
  4. Very specific:

    If user accepts: git merge --no-commit wip/[branch] then delete the wip branch

    line 28
  5. Distinctive:

    If user declines: continue to Phase 1

    line 29
  6. Very specific:

    Check ~/.claude/sessions/active-changes.log and ~/.claude/sessions/active-changes-*.log

    line 33
  7. Very specific:

    If no logs found, note: "No session hooks detected.

    line 34
  8. Distinctive:

    No checkpoints, no logs, no JSONs: "No crashed sessions.

    line 69
  9. Distinctive:

    Not a git repo: skip file status analysis

    line 72

“Common” is measured: kept by at least 6 in 10 of the 11 edited versions of this file we could read. “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

Recover Session

Recover work from a session that ended unexpectedly.

NOTE: For resuming planned features (with progress files), use /resume instead.

Phase 0: Check checkpoints

  1. Look for checkpoint files in .lifecycle/checkpoints/
  2. If found, present cognitive state alongside recovery data — not just "these files changed" but "you were doing X because Y"

Phase 0.5: Check WIP branches

  1. Check for wip/ branches: git branch --list 'wip/*'
  2. If found, present:
    WIP safety branches found:
  wip/project-20260331-1445 (2h ago) — 3 files saved
  wip/project-20260331-1200 (5h ago) — 1 file saved
  3. Offer: "Restore latest WIP branch? (merges saved files into current branch)"
  4. If user accepts: git merge --no-commit wip/[branch] then delete the wip branch
  5. If user declines: continue to Phase 1

Phase 1: Check crash logs (active-changes.log)

  1. Check ~/.claude/sessions/active-changes.log and ~/.claude/sessions/active-changes-*.log
  2. If no logs found, note: "No session hooks detected. Crash log recovery requires hooks that write to ~/.claude/sessions/active-changes.log. See continuum README for setup." Then skip to Phase 2.
  3. Parse # SESSION headers for cwd, sid, started
  4. Extract unique EDIT/WRITE file paths and COMMIT history
  5. Extract last STATE: line
  6. Determine each file's state via git:
  • git status --porcelain -- <file> → ~ still modified
  • git log --oneline --since="<started>" -- <file> → + committed
  • Neither → x lost
  1. Present:
Crashed session recovered:
Feature: [name] | Task: [from checkpoint or STATE]

Checkpoint (if available):
Context: [what was happening]
Next: [what was planned next]

Files ([N] edits across [M] files):
  ~ src/api/routes.ts — still modified
  + src/middleware/auth.ts — committed (abc1234)
  x tests/auth.test.ts — lost

Resume from here, start fresh with this context, or discard?
  1. Resume → continue working, delete logs. Fresh → present as context. Discard → delete logs.

Phase 2: Legacy session JSON files

  1. Scan ~/.claude/sessions/session-*.json for "status": "active"
  2. Same recovery flow as Phase 1

Edge Cases

  • No checkpoints, no logs, no JSONs: "No crashed sessions. Did you mean /resume?"
  • Sessions older than 7 days: flag as "likely stale" but allow recovery
  • Log with no EDIT/WRITE entries: show task only, note "no file changes tracked"
  • Not a git repo: skip file status analysis

$ARGUMENTS

Download now (ZIP)

  • Other versions13 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 13

    1. jamillazarev/multica-ops★ 11same name
    2. conorbronsdon/agent-skills★ 10same name
    3. conorbronsdon/agent-workspace★ 2same name
    4. garbageek/claude-plugins★ 1same name
    5. jamillazarev/opsinist★ 1same name
    6. OfficialDiLi/skills★ 1same name
    7. AR-Toqi/Mirai_mart★ 0same name
    8. dincho-kostadinov/invest-tracker-mvp★ 0same name
    9. dincho-kostadinov/spec-driven-development-pipeline★ 0same name
    10. eliphazbouye/skills★ 0same name

    3 more not listed

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • Used alongsideProjects with it also use imprint, remember and 5 more.

    Other files people use alongside this one.

    Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
    • OfficialDiLiSkill

      imprint

      After building any UI component, extract the visual patterns that matter for consistency and save them to ui-registry.md. So every component built after this one matches what came before.

      Also in 7 projects that use this

    • LangChainSkill

      remember

      Review the current conversation and capture valuable knowledge — best practices, coding conventions, architecture decisions, workflows, and user feedback — into persistent memory (AGENTS.md) or reusable skills. Use when the user says: (1) remember this, (2) save what we learned, (3) update memory, (4) capture learnings.

      Also in 7 projects that use this

    • Harut8Skill

      architect

      Also in 8 projects that use this

    • AWSSkill

      map

      Map verified service capabilities against customer-provided control framework and threat model. Produces mapping-results.json, controls-catalog.md, and framework-mapping.md. Supports pluggable frameworks: CCMv4, NIST 800-53, CIS Benchmarks, ISO 27001, etc. Split into 4 focused sub-skills for better accuracy; this skill orchestrates them.

      Also in 3 projects that use this

    • EffortlessMetricsSkill

      feature

      Also in 3 projects that use this

    • SentrySkill

      review

      Fixture review skill.

      Also in 8 projects that use this

  • In this repository6 other files it gives its agents.

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