The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local development commands to gather context. Specifically, it runs git log --oneline -10 to retrieve commit history, and npm run test and npm run build to capture project metrics and status. These are standard developer operations but involve subprocess execution.
[PROMPT_INJECTION]: The skill has an indirect prompt injection vulnerability surface. * Ingestion points: The skill reads project documentation (context/PROJECT_STATE.md) and command output from git log. * Boundary markers: No explicit delimiters or instructions are used to distinguish between trusted instructions and untrusted data from the git history. * Capability inventory: The skill has the ability to write to the file system (updating PROJECT_STATE.md and archiving sessions) and execute shell commands. * Sanitization: There is no evidence of sanitization or filtering applied to external data (like commit messages) before it is processed by the agent.

sprint-sync