build
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill accepts arbitrary user input through the
$ARGUMENTSvariable and passes it directly into a sequence of automation skills (make-plan,do,babysit) to perform coding tasks. - Ingestion points: Untrusted data enters the agent context via the
$ARGUMENTSplaceholder inSKILL.md. - Boundary markers: The skill lacks delimiters (such as XML tags or triple backticks) and explicit instructions to the agent to disregard embedded commands within the user input.
- Capability inventory: The skill is designed to "fully automate a coding session," which implies that the sub-skills it calls likely possess significant capabilities, such as writing to the file system or executing shell commands.
- Sanitization: There is no evidence of input validation, escaping, or filtering to prevent the interpolation of malicious instructions.
Audit Metadata