fittings-pkg-normalize-manifest
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use
curlfor interacting with the Fittings API to perform manifest normalization and catalog lookups.\n- [DATA_EXFILTRATION]: The skill transmits the contents ofpackage.jsonfiles to the external endpointhttps://fittings.sh/v1/pkg/normalize-manifestfor processing.\n- [PROMPT_INJECTION]: The skill processes untrustedpackage.jsondata, creating an indirect prompt injection surface where instructions could be embedded in manifest fields.\n - Ingestion points: The
manifestobject in the normalization request defined inSKILL.md.\n - Boundary markers: Absent; no instructions are provided to the agent to ignore or delimit embedded content.\n
- Capability inventory:
curlexecution for API requests.\n - Sanitization: Absent; no filtering or escaping of manifest text is performed before processing.\n- [CREDENTIALS_UNSAFE]: The documentation describes the use of the
FITTINGS_PRIVATE_KEYenvironment variable for payment signing, which results in sensitive cryptographic keys being stored in the agent's environment.
Audit Metadata