agentleFS
Sign inStart free
kimbjunnkimbjunn/session-pack/pack/SKILL.md

pack

Skill1 starsChanged 2 months agoMITView on GitHub

Manual-only entry point for the session-pack skill's packing workflow, triggered by the user typing `/pack`. Not for model auto-invocation — use the session-pack skill for that.

3 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions3 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 3

    1. getsentry/sentry-unreal★ 176same name
    2. ProjectAJ14/eklavya★ 29same name
    3. bradnovation/big-brains★ 1same name

    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.

  • In this repository2 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 0

    Distinctive

    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:

    Read <skill-dir>\SKILL.md in full (if not already in context) and follow its

    line 21
  2. Very specific:

    If the user supplied an argument with /pack, treat it as the session id (or a recognizable fragment/name) to pack.

    line 23
  3. Very specific:

    If they pick token-saving, --lean-level light|balanced|max tunes how hard it trims (balanced is the default).

    line 37

“Common” is measured: kept by at least 6 in 10 of the 3 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

/pack

Thin entry point. All logic, the shared script, and the reference material live in the session-pack skill. Do not reimplement anything here.

Locate that skill's directory first — do not assume a fixed path, since session-pack may be installed as a standalone skill (~/.claude/skills/session-pack or <repo>/.claude/skills/session-pack) or as a plugin ($env:USERPROFILE\.claude\plugins\marketplaces\session-pack, with ${CLAUDE_PLUGIN_ROOT} pointing at it). Call this resolved directory <skill-dir> below.

Steps

  1. Read <skill-dir>\SKILL.md in full (if not already in context) and follow its "Packing (sender)" section, steps 1-4, exactly.
  2. If the user supplied an argument with /pack, treat it as the session id (or a recognizable fragment/name) to pack. Otherwise do step 1 of that section: list candidate sessions by recency and ask which one.

  3. Ask which bundle they want — full fidelity or token-saving — unless the user already said (/pack --lean, "token-saving", "lean", "lightweight", "full fidelity", "as is"):

    • full fidelity (default): every byte of the transcript travels. Right when the record itself matters — an audit, an incident review, or a conversation that will later be synced into a repository.
    • token-saving (--lean): file bodies, long tool results, settled reasoning and images are replaced by a marked gap, so resuming costs the receiver roughly half the context (level-dependent; the command prints the measured figure). The conversation, the message tree and every tool call survive; it resumes normally. Right when the receiver is continuing the work, which is most handoffs.

    If they pick token-saving, --lean-level light|balanced|max tunes how hard it trims (balanced is the default). See session-pack's SKILL.md for the full flag table.

  4. Invoke the bundled script by absolute path:
    python <skill-dir>\scripts\ccpack.py pack <sessionId> -o <name>.ccpack --with-files [--lean]
    Pick the remaining flags (--with-files, --deep-files, --max-files, --with-history, --no-generic) per the guidance in session-pack's SKILL.md.
  5. Show the user REPORT.md and its slot table before letting the bundle leave the machine — this review step is mandatory, not optional.
  6. Report packed/missing/skipped/truncated file counts, and flag any missing files explicitly (the receiver will hit the same gap). With --lean, also relay the context ~X -> ~Y tokens (-Z%) line and say plainly that the bundle is abridged — the receiver sees the same notice in inspect.

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.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

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