foundry-spec

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for the agent to execute arbitrary shell commands as part of the implementation verification process. Specifically, 'verify' nodes with the 'run-tests' type are intended to run test suites directly via Bash, allowing for the execution of commands like 'pytest' or 'npm test' (documented in references/task-hierarchy.md and references/modification-operations.md).
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to its core function of analyzing external codebase content.
  • Ingestion points: The skill uses tools such as Grep, Read, and LSP-based analysis to ingest content from the project's files (SKILL.md, references/codebase-analysis.md).
  • Boundary markers: No explicit delimiters or instructions to treat codebase content as data rather than instructions are provided in the prompts.
  • Capability inventory: The agent can write/modify files and execute shell commands (references/task-hierarchy.md).
  • Sanitization: The skill lacks documented sanitization or validation of the data extracted from the codebase before it is passed to AI models for planning or review tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 12:39 AM