stelow-workflow-testing-ai-code
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from project specification files and interpolates the results into shell commands and logic.
- Ingestion points: Reads product metadata, appetite, and scope from
spec-product.md,spec-tech.md, and project files likepackage.jsonorrequirements.txt(SKILL.md). - Boundary markers: Absent. The skill does not use delimiters or instructions to ignore embedded content within the ingested specification files.
- Capability inventory: Executes subprocesses for
npm,pytest,cargo,find,rg,jq, andmadgeacross multiple testing and analysis steps (SKILL.md). - Sanitization: Absent. There is no evidence of input validation or escaping for data extracted from project files before it is used in shell command construction.
- [COMMAND_EXECUTION]: The skill utilizes the agent's ability to run shell commands to perform tech stack detection, dependency analysis, and regression testing.
- Evidence: Instructs the agent to run complex pipelines such as
find . -path ./node_modules -prune -o -name "*.test.*" -print | xargs rg -l "module_name" > affected_tests.txtandnpm test -- --testPathPattern="$(cat affected_tests.txt | tr '\n' '|')"(SKILL.md).
Audit Metadata