agentleFS
EthanLLLEthanLLL/loom/skills/profiles/SKILL.md

profiles

Skill8 starsChanged 4 days agoMITView on GitHub

Add, change or remove a Claude Code profile — another model provider (an API key with a base URL, an Anthropic key, a different Bedrock setup) that a new Loom session can be started on. Use when the user wants a session to use a different key, endpoint, relay or provider ("加个 profile", "用这个 key", "base url", "中转", "API key", "换个模型供应商").

  • Reads credentials

4 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions4 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 4

    1. alxyrgin/health-os★ 38same name
    2. ayrshare/ayrshare-social-media-api-claude-plugin★ 19same name
    3. peterbamuhigire/proposal-skills★ 11same name
    4. postproxy/postproxy-claude-plugin★ 0same 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 repository6 other files it gives its agents.

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. Very specific:

    Every file there shows up in New chat as "Claude Code · <name>".

    line 9
  2. Distinctive:

    Run one prompt through it.

    line 82

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

profiles: which provider a Claude Code session talks to

A profile is a Claude Code settings file at ~/.loom/profiles/<name>.json. Every file there shows up in New chat as "Claude Code · <name>"; a session started on it passes the file to the CLI as --settings, so its env overrides the machine's ~/.claude/settings.json. The choice is fixed for the session.

  • <name>: letters, digits, -, _, .; at most 64 characters.
  • The file holds a key: create it with mode 0600, and never print the key back.
  • Loom reads the file at each spawn. An edit applies to a running session the next time its process starts (after an idle release, or /reload).
  • Removing a file breaks the sessions on it: their turns fail with "profile <name> not found". Check grep -l '"cc_profile":"<name>"' ~/.loom/sessions/*/meta.json first and tell the user which ones.
  • The composer's model menu (Opus / Sonnet / Haiku) picks between ANTHROPIC_DEFAULT_OPUS_MODEL, _SONNET_ and _HAIKU_. Set all three to names the endpoint knows. One left out falls back to the machine's id for that alias, which another provider will refuse.

An endpoint with a key (Anthropic-compatible: relays, LiteLLM, Anthropic API)

{
  "env": {
    "CLAUDE_CODE_USE_BEDROCK": "0",
    "ANTHROPIC_BASE_URL": "https://example.com",
    "ANTHROPIC_AUTH_TOKEN": "sk-…",
    "ANTHROPIC_DEFAULT_OPUS_MODEL": "<model name the endpoint expects>",
    "ANTHROPIC_DEFAULT_SONNET_MODEL": "<…>",
    "ANTHROPIC_DEFAULT_HAIKU_MODEL": "<…>"
  }
}
  • ANTHROPIC_AUTH_TOKEN is sent as Authorization: Bearer; an endpoint that wants x-api-key takes ANTHROPIC_API_KEY instead.
  • The base URL is without /v1: CC appends /v1/messages.
  • Set all three model names: the machine's defaults are Bedrock ids, which another endpoint will not know. Ask the user for the names if the provider's docs do not say.
  • Set "CLAUDE_CODE_USE_BEDROCK": "0" even if this machine does not use Bedrock today; it costs nothing.

Bedrock API keys (bedrock-api-key-…)

Bedrock's Anthropic-compatible endpoint takes them as a bearer token, with two more lines than above (measured, CC 2.1.300):

{
  "env": {
    "CLAUDE_CODE_USE_BEDROCK": "0",
    "ANTHROPIC_BASE_URL": "https://bedrock-runtime.<region>.amazonaws.com/anthropic",
    "ANTHROPIC_AUTH_TOKEN": "bedrock-api-key-…",
    "ANTHROPIC_DEFAULT_OPUS_MODEL": "us.anthropic.claude-opus-5-5",
    "ANTHROPIC_DEFAULT_SONNET_MODEL": "us.anthropic.claude-sonnet-5-5",
    "ANTHROPIC_DEFAULT_HAIKU_MODEL": "us.anthropic.claude-haiku-5-5",
    "CLAUDE_CODE_EXTRA_BODY": "{\"metadata\":{\"user_id\":\"loom\"}}",
    "CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS": "1"
  }
}
  • Model ids are inference-profile ids (us./eu./global. prefix). A bare anthropic.… id is refused for on-demand use, claude-opus-5-5 is invalid; no [1m] suffix (the window is 1M anyway).
  • Without CLAUDE_CODE_EXTRA_BODY: "400 Request metadata contains a value that violates the regular expression". CC sends metadata.user_id as a JSON string; the extra body replaces it.
  • Without CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS: "400 Unexpected value(s) … for the anthropic-beta header".

Other relays can fail the same two ways; the same two lines fix them.

Check it before handing it over

Run one prompt through it; a wrong URL, key or model name shows up here rather than in the user's first turn:

claude -p --setting-sources user --settings ~/.loom/profiles/<name>.json 'Reply with OK.'

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.