payment-failure-recovery-advisor
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a basic Python scaffold for processing JSON data. It reads an input file and writes a summary to an output file based on command-line arguments.
- [INDIRECT_PROMPT_INJECTION]: The skill processes structured input (JSON/CSV), which is an ingestion point for untrusted data. However, the reference implementation in
run.pyonly performs a schema-safe operation (listing keys) and lacks any dangerous capabilities like network access, shell execution, or sensitive file manipulation that could be exploited via injection. - Ingestion points:
run.pyreads data from a path specified by the--inputargument. - Boundary markers: None identified.
- Capability inventory: Basic file system read and write restricted to local paths provided at runtime.
- Sanitization: Uses standard
json.loads()for parsing.
Audit Metadata