agentleFS
Sign inStart free
ShurikenTradeShurikenTrade/shuriken-skills/skills/scoping/SKILL.md

scoping

Skill90 starsChanged 7 months agoMITView on GitHub

Use when a user asks what an agent key can do, how Shuriken permissions work, or which scopes they need for a given integration.

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. NashEQify/forge★ 2same name
    2. augments-labs/sdlc-skills★ 1same name
    3. Yeseh/software-factory-demo★ 1same name
    4. smileynet/game-spice★ 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 repository4 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

8 rules. 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 8

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every agent key carries a set of scopes.

    line 8
  2. Distinctive:

    Start from what the integration does on its hot path, and enumerate only those capabilities.

    line 16
  3. Distinctive:

    When in doubt, grant less and widen later if the integration errors on a missing scope.

    line 17
  4. Distinctive:

    Create a separate agent key per integration (or per purpose) rather than one fat key reused everywhere — that way each key's scope surface …

    line 18
  5. Distinctive:

    If a scope is required for a one-off setup step but not for steady-state operation, consider doing the setup with a broader key and then …

    line 19
  6. Distinctive:

    Separate read from write: Read scopes are generally safe.

    line 25
  7. Distinctive:

    If the scope is genuinely needed, guide the user to rotate the key with the added scope.

    line 49
  8. Distinctive:

    If the scope should not be needed (the call is read-only but erroring on a write scope), something in the integration is calling the wrong …

    line 50

“Common” is measured: kept by at least 6 in 10 of the 4 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

Scoping an agent key

Every agent key carries a set of scopes. A scope is a capability grant — read tokens, execute trades, read positions, deliver notifications. The server enforces scopes on every tool call and API endpoint; a call outside the granted scope fails with a structured authorization error.

Best practice: grant the minimum scopes the agent key actually needs

This is the single most important rule for agent-key scoping. For every agent key, grant only the scopes the integration actively uses — nothing speculative, nothing "just in case," nothing broader than the job requires.

Concretely:

  • Start from what the integration does on its hot path, and enumerate only those capabilities.
  • When in doubt, grant less and widen later if the integration errors on a missing scope.
  • Create a separate agent key per integration (or per purpose) rather than one fat key reused everywhere — that way each key's scope surface stays tight.
  • If a scope is required for a one-off setup step but not for steady-state operation, consider doing the setup with a broader key and then running with a narrower one.

Broader scopes are a liability: a leaked key is only as dangerous as the capabilities it carries.

Principles

  • Separate read from write. Read scopes are generally safe; write/execute scopes move money or send messages and deserve more scrutiny.
  • Per-integration, not per-user-role. A scope is a capability, not a persona. Reason about what the integration does, not who the user is.

Scope categories

Authoritative scope names live in the docs; these are the categories to reason about:

  • Read scopes — querying token info, positions, wallet balances, alpha signals. Safe for analytics and display use cases.
  • Trading scopes — planning trades, executing trades, cancelling orders. Required for any integration that moves funds.
  • Delivery scopes — sending notifications (Telegram, email) on behalf of the user. Rate-limited.
  • Administrative scopes — managing the user's agent keys themselves, linked wallets, etc. Typically not needed by third-party integrations.

Fetch https://docs.shuriken.trade/llms.txt for the current authoritative scope list and the exact names to pass when minting a key.

Common scope combinations

  • Read-only analytics dashboard: read scopes only.
  • Automated trading bot: read scopes + trading scopes.
  • Notification/alert service: read scopes + delivery scopes.
  • Full-service trading + notifications: read scopes + trading scopes + delivery scopes.

When a call fails on scope

The error response names the missing scope. Two valid responses:

  1. If the scope is genuinely needed, guide the user to rotate the key with the added scope.
  2. If the scope should not be needed (the call is read-only but erroring on a write scope), something in the integration is calling the wrong endpoint — investigate before granting.

Pointers

  • Platform documentation (including scope reference): fetch https://docs.shuriken.trade/llms.txt and search for "scopes"
  • Related skills: shuriken:agent-keys, shuriken:api-integration

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