The Agent Skills Directory

Prompt Injection (SAFE): No instructions found that attempt to override system safety guidelines or extract system prompts. The instructions are strictly limited to the task of updating the exercise database.
Data Exposure & Exfiltration (SAFE): The skill interacts with a local source file src/data/popularExercises.ts. No access to sensitive files (like .env or SSH keys) or network exfiltration patterns were detected.
Unverifiable Dependencies & Remote Code Execution (SAFE): The skill uses local commands grep and pnpm type-check. It does not download or execute scripts from remote or untrusted sources.
Indirect Prompt Injection (LOW):
Ingestion points: User input triggers such as "add exercise" or "new exercise".
Boundary markers: Absent. The skill does not explicitly instruct the agent to ignore embedded instructions in exercise names.
Capability inventory: File system write (src/data/popularExercises.ts) and command execution (pnpm type-check).
Sanitization: Absent. The skill relies on the agent correctly formatting the TypeScript object based on user-provided strings. While a user could attempt to provide a name containing code, the rigid schema and subsequent type-check provide basic guardrails.

add-exercises