agentleFS
Sign inStart free
nuffinnuffin/hermes-task-framework/skills/task-lifecycle-portability/SKILL.md

task-lifecycle-portability

Skill2 starsChanged 45 days agoMITView on GitHub

Export, import, and rebuild self-contained task snapshots without mirrors or symlinks.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before reporting export success:

    line 45
  2. Distinctive:

    Run task artifact audit.

    line 47
  3. Distinctive:

    Verify task identity/hash and complete context after extraction.

    line 57
  4. Very specific:

    Use this skill for one-task/offline transfer.

    line 64

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Task Lifecycle Portability

Owns one-task snapshot transport. Task creation, status, rename, retirement, and deletion belong to the lifecycle layer; damaged-file reconstruction belongs to task-lifecycle-edge-cases.

Storage contract

The task directory is the only source of truth. Canonical files are real files; there is no per-profile/per-hash mirror, symlink protection, or relink command.

  • input/ contains source material and is never cleaned.
  • output/ contains generated evidence and is cleared only by explicit hard reset.
  • Root MEMORY/CHANGELOG and memories/ are persistent context.

Commands

From the task-framework skill directory:

python3 scripts/manage_task.py export <hash-or-dir>
python3 scripts/manage_task.py import <archive.tar.gz-or-zip>
python3 scripts/manage_task.py rebuild <hash>

Export contract

Current export creates a self-contained tar.gz beside the task directory and includes the complete task except .git, including input/, output/, scripts, metadata, and hierarchical context. Complete evidence is the safe default; output exclusion requires a future explicit, tested option and is not implied.

Before reporting export success:

  1. Run task artifact audit.
  2. Confirm archive members are contained under one task directory.
  3. Confirm canonical files and context pairs are present.
  4. Record archive path and checksum.

Import/rebuild contract

  • Accept tar.gz or zip only after archive path traversal checks.
  • Refuse overwrite when the destination task exists.
  • Restore real files, never symlinks or mirrors.
  • Verify task identity/hash and complete context after extraction.
  • Regenerate indexes after successful import.

rebuild locates a matching snapshot and delegates to import. Ambiguous snapshots must be reported rather than silently selecting the wrong task.

Cross-machine choice

Use this skill for one-task/offline transfer. Use task-cross-machine-sync for an explicitly configured shared Git task root.

See references/output-model-design.md for the canonical input/output boundary.

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