The Agent Skills Directory

[DATA_EXFILTRATION]: The skill performs network operations using curl to communicate with api.firepan.com. It transmits repository URLs and requires the handling of sensitive Authorization: Bearer tokens for authenticated calls.
[COMMAND_EXECUTION]: The instructions include shell and Python examples for generating idempotency keys and executing network requests, which involves the use of system utilities like curl and uuidgen.
[PROMPT_INJECTION]: The skill is subject to indirect prompt injection as it processes untrusted data from external smart contract repositories.
Ingestion points: Smart contract code is retrieved from external sources via the repo_url parameter defined in SKILL.md.
Boundary markers: There are no explicit instructions or delimiters provided to prevent the agent from obeying instructions embedded within the contract code being analyzed.
Capability inventory: The skill possesses network capabilities (curl) to send data to external endpoints.
Sanitization: No sanitization or validation logic is specified for the contract content before it is processed by the agent or the API.

smart-contract-audit