The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) as it processes content from external story files to define its validation workflow.
Ingestion points: Markdown files located in docs/stories/ are read in Phase 1 to extract acceptance criteria and development notes.
Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are used when processing the content of these story files.
Capability inventory: The agent has permissions to execute shell commands (test runners), write new documentation, and edit existing files across the repository.
Sanitization: The skill does not implement sanitization or validation of the extracted criteria before using them to guide its logic.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute standard test runners, including make, pytest, npm, and bun. These operations are limited to test-related commands as defined in the skill's tool constraints, aligning with the skill's primary purpose of code validation.

forge-qa