The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill defines a dependency on another skill hosted in the vendor's repository: https://github.com/BankrBot/openclaw-skills/blob/main/bankr/SKILL.md.
[COMMAND_EXECUTION]: The skill employs curl to interact with the Bankr API and the mining coordinator, and uses system tools like jq, openssl, and uuidgen for processing JSON responses and generating nonces.
[CREDENTIALS_UNSAFE]: The skill requires a BANKR_API_KEY for authentication. Instructions correctly guide users to set this as an environment variable and specify the need for write access to enable transaction signing, which is essential for the skill's primary function.
[PROMPT_INJECTION]: The skill identifies a potential attack surface for indirect prompt injection via challenge data retrieved from the mining coordinator.
Ingestion points: Challenge documents (doc) and specific solving instructions (solveInstructions) provided by coordinator.agentmoney.net.
Boundary markers: The instructions include a explicit warning to treat external payloads as challenge data rather than trusted system instructions.
Capability inventory: The skill can perform wallet balance checks, token swaps, and on-chain transaction submissions through the Bankr API.
Sanitization: The skill instructs the agent to review all external response content before acting on it to prevent unauthorized behavior changes.

botcoin-miner