spec-driven-workflow
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes local Python scripts (
spec_validator.py,test_extractor.py,spec_generator.py) to automate parts of the development workflow, such as generating templates and extracting test cases from markdown files. These scripts rely on the Python standard library and do not perform network operations. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided specification files (
spec.md). This creates a potential surface for indirect prompt injection where instructions embedded in the user's data could influence the agent's behavior during the validation or testing phases. - Ingestion points: Files are read in Phase 3 (validation) and Phase 4 (test extraction) via the
--fileargument. - Boundary markers: The instructions do not define explicit boundary markers or delimiters to isolate user data from the agent's core instructions.
- Capability inventory: The skill includes scripts capable of reading from and writing to the local file system using the
pathlibmodule. It does not possess network access or dynamic execution capabilities (exec/eval). - Sanitization: The provided scripts use regular expressions for pattern matching and text extraction but do not perform specific sanitization of input text to prevent prompt injection attacks.
Audit Metadata