agentleFS
MoeenNehzatiMoeenNehzati/famulus/skills/cloud-files/SKILL.md

cloud-files

Skill4 starsChanged 4 months agoMITView on GitHub

Use when the user or another skill needs to read from or write to the configured LLM root of a remote. Do not use for local files or remote paths outside that LLM root.

  • Reads credentials

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 are 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.

  • 2

    Very specific

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

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

    When first exposed to this skill in a session, invoke famulus_dispatcher.invoke_security_2 once with:

    line 12
  2. Distinctive:

    Do not repeat this initial call during the session.

    line 27
  3. Distinctive:

    When this skill is used, begin with:

    line 94
  4. Very specific:

    Use lists-read, lists-write, lists-delete for routine list storage and plans-read, plans-write, plans-delete for plan storage within their …

    line 117
  5. Distinctive:

    If a script exits nonzero, report the visible error and do not infer remote state beyond what the successful output established.

    line 125

≈ “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
<!-- BEGIN BLUEPRINT INTERFACES -->

Generated from blueprint.yaml. Do not edit this block by hand.

Managed setup

When first exposed to this skill in a session, invoke famulus_dispatcher.invoke_security_2 once with:

{
  "caller": "cloud-files",
  "interface": "cloud-files.interface.setup",
  "version": 2,
  "arguments": {
    "positionals": [],
    "options": {},
    "stdin": null
  }
}

Do not repeat this initial call during the session. Obtain permission before carrying out setup, then follow the returned setup-manager instructions exactly. If the result is busy or failed, stop and report it.

Executable Interfaces:

Send the required caller (caller skill), interface, version, and arguments; optional dry_run defaults to false. Compact uses ordered positionals plus an option mapping; ordered raw argv uses positionals: [] plus every argv token in list options. Never mix forms.

  • cloud-files._rtx.interface.lists-delete — Delete a file from cloud storage under the lists/ directory.
    • Caller: cloud-files
    • Version: 1
    • Security level: 2
    • Alternative: default Arguments JSON (replace labels with actual values). Omit optional positionals and options that are not needed. {"options": {}, "positionals": ["lists/<path>"], "stdin": null} Required options: []; positional arity: 1..1; stdin: forbidden
  • cloud-files._rtx.interface.lists-read — Read a file from cloud storage under the lists/ directory.
    • Caller: cloud-files
    • Version: 1
    • Security level: 2
    • Alternative: default Arguments JSON (replace labels with actual values). Omit optional positionals and options that are not needed. {"options": {}, "positionals": ["lists/<path>"], "stdin": null} Required options: []; positional arity: 1..1; stdin: forbidden
  • cloud-files._rtx.interface.lists-write — Write content (from stdin) to a file in cloud storage under the lists/ directory.
    • Caller: cloud-files
    • Version: 1
    • Security level: 2
    • Alternative: default Arguments JSON (replace labels with actual values). Omit optional positionals and options that are not needed. {"options": {}, "positionals": ["lists/<path>"], "stdin": null} Required options: []; positional arity: 1..1; stdin: permitted
  • cloud-files._rtx.interface.plans-delete — Delete a file from cloud storage under the plans/ directory.
    • Caller: cloud-files
    • Version: 1
    • Security level: 2
    • Alternative: default Arguments JSON (replace labels with actual values). Omit optional positionals and options that are not needed. {"options": {}, "positionals": ["plans/<path>"], "stdin": null} Required options: []; positional arity: 1..1; stdin: forbidden
  • cloud-files._rtx.interface.plans-read — Read a file from cloud storage under the plans/ directory.
    • Caller: cloud-files
    • Version: 1
    • Security level: 2
    • Alternative: default Arguments JSON (replace labels with actual values). Omit optional positionals and options that are not needed. {"options": {}, "positionals": ["plans/<path>"], "stdin": null} Required options: []; positional arity: 1..1; stdin: forbidden
  • cloud-files._rtx.interface.plans-write — Write content (from stdin) to a file in cloud storage under the plans/ directory.
    • Caller: cloud-files
    • Version: 1
    • Security level: 2
    • Alternative: default Arguments JSON (replace labels with actual values). Omit optional positionals and options that are not needed. {"options": {}, "positionals": ["plans/<path>"], "stdin": null} Required options: []; positional arity: 1..1; stdin: permitted
  • cloud-files._rtx.interface.write-config — Write <CONFIG>/cloud-files/config.json with the given remote LLM root. Owned by cloud-files.
    • Caller: cloud-files
    • Version: 1
    • Security level: 2
    • Alternative: default Arguments JSON (replace labels with actual values). Omit optional positionals and options that are not needed. {"options": {"--dry-run": true, "--home": "dir", "--remote-llm-root": "path"}, "positionals": [], "stdin": null} Required options: ["--home"]; positional arity: 0..0; stdin: forbidden

Instruction Interfaces:

These are LLM-readable instruction surfaces. Read and follow them directly; do not invoke the MCP server for them.

  • connect-google.interface.default@1 — Route Google OAuth-client preparation according to whether a valid Desktop client is already installed.
<!-- END BLUEPRINT INTERFACES -->

When this skill is used, begin with:

Skill: cloud-files

0. Boundary

This skill owns Google Drive transport. Other skills should call this skill's scripts rather than speaking to the Drive API directly.

Install-time config lives at <CONFIG>/cloud-files/config.json. Legacy OAuth credentials live at <CONFIG>/cloud-files/credentials.json.

For shared Google setup or Drive reauthorization, first invoke connect-google.interface.default. Its deterministic coordinator creates a credential file, asks Drive's owner to probe live access, and stores the path only after verification. Treat only complete: true as successful setup; report an incomplete Drive result and retry through connect-google with the same file.

Existing legacy Drive credentials remain runtime-readable until a verified credential-file binding replaces them. Do not offer legacy setup as a new route.

1. Preapproved LLM-root operations

Use lists-read, lists-write, lists-delete for routine list storage and plans-read, plans-write, plans-delete for plan storage within their respective directories.

Each interface takes a single positional path argument constrained to its directory prefix (lists/ or plans/). Write interfaces read file content from stdin.

2. Separately prompted broader reads

A broader read from the Google Drive root is available via a script not registered as a dispatcher interface. It is intentionally not listed in blueprint.yaml:suggested_permissions.

If a script exits nonzero, report the visible error and do not infer remote state beyond what the successful output established.

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.