agentleFS
System-Error-WorldwideSystem-Error-Worldwide/the-loop/.agents/skills/the-loop-setup/SKILL.md

the-loop-setup

Skill0 starsChanged 3 months agoMITView on GitHub

Detect supported agent harnesses, preview an exact portable-skill installation, and apply or roll back only approved filesystem operations. Use when a user asks to install, update, inspect an install plan for, or uninstall SYSTEM ERROR'S THE LOOP.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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:

    Read the four public adapter manifests and detect installed executables independently.

    line 26
  2. Very specific:

    Write the schema-valid receipt to .the-loop/installs/<receipt-id>.json.

    line 31

≈ “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

THE LOOP Setup

Install the public skill pack into one repository or one selected user scope. Start with a dry run. Never infer approval from a broad request when the plan reports a collision.

Inputs

  • Source root containing canonical packages under .agents/skills: either the public checkout or an installed .the-loop/toolkit.
  • Existing target root.
  • One or more installed harnesses, or automatic detection.
  • Scope: project or user.
  • Mode: supported copy. A requested link plan fails closed when portable documentation transformation requires copies; unchanged synthetic packages exercise the lower-level link path only with explicit same-filesystem proof.
  • Exact destination approvals for every non-identical collision.

Procedure

  1. Read the four public adapter manifests and detect installed executables independently.
  2. Build a read-only operation plan in stable order. Include directory creation, copy, link, skip, collision, rollback action, and digests.
  3. Show the plan. A dry run changes nothing.
  4. On apply, recheck target identity, source digests, destination digests, selected harnesses, mode, and exact approvals.
  5. Apply the transaction. Restore the exact prior filesystem state if any operation or receipt write fails.
  6. Write the schema-valid receipt to .the-loop/installs/<receipt-id>.json.
  7. For rollback, remove only unchanged receipt-owned output. Restore an approved replacement only when the installed result is unchanged and its receipt backup is present.

Apply installs a private, namespaced .the-loop/toolkit containing the canonical packages, adapters, protocols, schemas, scripts and standard-library runtime. After the first apply, core execution, Doctor, rollback and a later Setup plan do not require network access or the original source checkout.

From an installed project, use .the-loop/toolkit/scripts/the_loop_setup.py. The canonical public source is scripts/the_loop_setup.py, and the receipt contract is install-receipt.schema.json.

Evidence

  • Dry-run plan and selected adapter manifests.
  • Applied or rolled-back install receipt.
  • Setup tests covering collision refusal, exact approval, interruption, changed output, shipping-link rejection, and bounded unchanged-package link proof.

Halt conditions

Stop with a precise error when no selected harness is installed, an adapter is missing or invalid, a path escapes its root, any path crosses a symlink, a source or destination changed after planning, link support is not proven, or an exact collision approval is absent.

Setup does not read credential files, collect prompts, send telemetry, or publish anything.

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.