grill-with-docs
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is designed to execute three local Python scripts provided in its package (
scripts/context_md_linter.py,scripts/adr_scanner.py, andscripts/glossary_code_consistency.py) to automate the validation of documentation and consistency checks against the codebase. - [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection due to its core functionality of reading and processing untrusted data from the user's project files.
- Ingestion points: The scripts
adr_scanner.py,context_md_linter.py, andglossary_code_consistency.pyread and parse content from Markdown documentation (e.g.,CONTEXT.md,docs/adr/) and various source code files across the project directory. - Boundary markers: The instructions do not specify the use of delimiters or boundary markers to isolate the output of these analysis scripts, which may include content from the scanned files, from the agent's system instructions.
- Capability inventory: The skill facilitates the execution of Python scripts and recursive file reading throughout the codebase.
- Sanitization: The analysis scripts focus on structural validation and term frequency counting but do not implement sanitization or filtering to prevent malicious instruction patterns within the processed files from being interpreted as commands by the agent.
Audit Metadata