The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to read and process content from untrusted local files (e.g., docs/stories/**/*.md). An attacker with write access to the repository could attempt to embed malicious instructions within these stories. However, the skill provides a rigid framework for implementation and verification, which serves as a functional control.
Ingestion points: User story files are read from the repository using the Read tool.
Boundary markers: No explicit delimiters or XML tags are utilized to encapsulate content from story files.
Capability inventory: The agent possesses capabilities to modify files and execute shell commands (e.g., git, npm, pytest).
Sanitization: No explicit sanitization of story file content is described.
[COMMAND_EXECUTION]: The skill facilitates the execution of local development and verification commands such as make test, npm run lint, and pytest. These operations are integral to the skill's purpose as a software implementation assistant and are restricted to the tools specified in the platform configuration.

forge-dev