grill-me
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external documentation and source code which could contain adversarial instructions.
- Ingestion points: Documented in 'Phase 1: Read Context and Scope' and 'Phase 5: Explore the Codebase', where the skill uses 'Read', 'Grep', and 'Glob' to process design specs and existing source code.
- Boundary markers: The instructions do not specify the use of delimiters or specific instructions to ignore embedded commands within the processed project files.
- Capability inventory: The skill is permitted to use 'Bash', 'Write', and 'Edit' tools, providing a pathway for an indirect injection to influence the filesystem or execute shell commands.
- Sanitization: There is no logic provided to filter or validate the content of the documents or code before the agent incorporates them into its reasoning process.
Audit Metadata