throttling-apis
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes potentially untrusted data from the {baseDir}/api-specs/ directory, creating a vulnerability surface for indirect prompt injection.
- Ingestion points: The agent utilizes the Read tool to examine files in the {baseDir}/api-specs/ directory, as instructed in SKILL.md and references/implementation.md.
- Boundary markers: The skill's instructions do not define any delimiters or explicit safety instructions to ensure the agent ignores embedded commands within the ingested data.
- Capability inventory: The skill permits the use of Write, Edit, and Bash(api:throttle-*) tools, which represent significant capabilities that could be misdirected if an injection occurs.
- Sanitization: There are no steps provided for the validation or sanitization of the content found within the API specification files.
Audit Metadata