pattern-matching
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to perform a 'SURVEY' and 'STUDY' of existing files within a repository to replicate their structure, naming, and logic. This process of shadowing external content creates an attack surface for indirect prompt injection, where an attacker could embed malicious instructions in the source repository's comments or documentation that the agent might obey while attempting to conform to the project's style.
- Ingestion points: Existing repository files used as precedents for new code generation, as described in the 'Entry Protocol' and 'Shadowing Method' sections.
- Boundary markers: The instructions lack explicit boundary markers or directives for the agent to ignore embedded natural language instructions found within the code files it analyzes.
- Capability inventory: The skill assumes the underlying agent possesses file-reading capabilities to perform surveys and file-writing capabilities to replicate the code.
- Sanitization: There is no mention of sanitizing or filtering the ingested content. However, the skill does include a manual guardrail requiring the agent to 'Raise it with the human first' and states that 'Unilateral convention changes are forbidden,' which provides a level of human oversight.
Audit Metadata