agentleFS
worldlibertyworldliberty/agentpay-sdk/skills/agentpay-sdk/agents/CLAUDE.md

agentpay-sdk / agents

CLAUDE.md460 starsChanged 8 months agoMITView on GitHub

Use the installed agentpay-sdk skill whenever the task touches wallet setup, funding, spending limits, transfers, paid APIs, approvals, or backups. Rules that must stay consistent:

The file

Show its rules and text

We split this CLAUDE.md into the rules it gives an agent

14 rules. About 0 are common, 2 are distinctive, and 12 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 12

    Very specific

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

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

    Use the installed agentpay-sdk skill whenever the task touches wallet setup, funding, spending limits, transfers, paid APIs, approvals, or …

    line 3
  2. Very specific:

    use agentpay wallet --json to decide whether a reusable wallet already exists.

    line 8
  3. Very specific:

    if the user only asks what the skill can do, answer from SKILL.md and do not probe the machine first

    line 9
  4. Distinctive:

    never ask the user to paste a wallet backup passphrase, agent auth token, or paying-proxy URL into chat.

    line 10
  5. Very specific:

    never run sudo or agentpay admin decisions for the user.

    line 11
  6. Very specific:

    if AgentPay is not set up and the user is trying to use the wallet, tell them to run agentpay setup locally

    line 12
  7. Very specific:

    if the wallet exists and the user wants to preserve it while re-running setup, tell them to run agentpay setup locally and choose "Keep the …

    line 13
  8. Very specific:

    if the local wallet is gone but the user has an encrypted backup, tell them to run agentpay setup locally and choose "Restore from a backup"

    line 14
  9. Very specific:

    after a fresh setup, tell the user to create an encrypted offline backup with agentpay admin wallet-backup export --output <PATH> unless …

    line 15
  10. Very specific:

    if the user does not specify network or asset for a payment, default to USD1 on bsc

    line 16
  11. Distinctive:

    check funding before outbound actions

    line 17
  12. Very specific:

    if funding is missing, stop and ask the user to fund the wallet with the exact address, network, token needs, and BNB gas needs (agentpay …

    line 18
  13. Very specific:

    use agentpay --help and subcommand help as the source of truth for exact flags

    line 23
  14. Very specific:

    use agentpay admin limits (run by the user) as the default path for spending limit changes

    line 24

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md 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
CLAUDE.md

AgentPay SDK Adapter

Use the installed agentpay-sdk skill whenever the task touches wallet setup, funding, spending limits, transfers, paid APIs, approvals, or backups.

Rules that must stay consistent:

  • start from agentpay config show --json
  • use agentpay wallet --json to decide whether a reusable wallet already exists; exit code 4 means AgentPay is not set up for this user
  • if the user only asks what the skill can do, answer from SKILL.md and do not probe the machine first
  • never ask the user to paste a wallet backup passphrase, agent auth token, or paying-proxy URL into chat; there is no vault password
  • never run sudo or agentpay admin decisions for the user; admin actions need the user's Touch ID, Mac password, or sudo, so give them the exact command to run locally
  • if AgentPay is not set up and the user is trying to use the wallet, tell them to run agentpay setup locally
  • if the wallet exists and the user wants to preserve it while re-running setup, tell them to run agentpay setup locally and choose "Keep the existing wallet"
  • if the local wallet is gone but the user has an encrypted backup, tell them to run agentpay setup locally and choose "Restore from a backup"
  • after a fresh setup, tell the user to create an encrypted offline backup with agentpay admin wallet-backup export --output <PATH> unless they already have a verified backup
  • if the user does not specify network or asset for a payment, default to USD1 on bsc
  • check funding before outbound actions
  • if funding is missing, stop and ask the user to fund the wallet with the exact address, network, token needs, and BNB gas needs (agentpay fund --json)
  • pass --json and read exit codes: 0 ok, 1 failure (follow error.next_action), 2 usage, 3 waiting for manual approval, 4 not set up
  • amounts for transfer, transfer-native, approve, and the Solana transfers are decimal whole units; amounts for x402 and mpp are integer base units
  • if a request hits manual approval, say it is waiting for the user's approval and give them agentpay admin approve <UUID>; for EVM and Solana --broadcast payments, tell them to keep the original command running and not rerun it; if that command already exited with code 3, use agentpay resume --approval-request-id <UUID> after approval
  • Stripe Link, Bitrefill, and private transfers are not available
  • use agentpay --help and subcommand help as the source of truth for exact flags
  • use agentpay admin limits (run by the user) as the default path for spending limit changes

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.