koda-plan
Pass
Audited by Gen Agent Trust Hub on Jul 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it incorporates untrusted data from project documentation and code comments into the agent's planning and execution process.- Ingestion points: The skill reads user requests, project file structures, and documentation or comments (SKILL.md, Sections 1.1, 1.3, and 1.4).- Boundary markers: The skill does not define delimiters or specific instructions to isolate data read from files from the agent's core instructions.- Capability inventory: The skill possesses file-write capabilities (saving plans to
.koda/plans/) and the ability to invoke other available agent skills and tools (SKILL.md, Sections 2.4 and 3.1).- Sanitization: No sanitization, filtering, or validation of the content ingested from external project files is specified.
Audit Metadata