agile-skill-build
Pass
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill scripts scaffold.py and build.py execute local file system operations to create directory structures and write the assembled AGENTS.md output file.
- [COMMAND_EXECUTION]: The Python scripts dynamically modify the search path using sys.path.insert to load the engine module from a relative sibling directory (../abd-shaping/scripts). This enables the skill to utilize centralized logic provided by the author.
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as build.py assembles content from markdown files. Ingestion points: build.py reads files in the content/ directory. Boundary markers: None are used to separate ingested files. Capability inventory: build.py and scaffold.py have file system write access. Sanitization: No validation or escaping is performed on the ingested markdown content before it is merged into the final instructions.
Audit Metadata