The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to verify backend code by running npx tsc --noEmit and npx convex dev --once. These are standard development workflows for the target platform using official tools.- [EXTERNAL_DOWNLOADS]: Recommends utilizing libraries such as @convex-dev/agent and @convex-dev/workflow. These are official packages associated with the platform vendor.- [INDIRECT_PROMPT_INJECTION]: The workflow requires the agent to ingest external project files, specifically convex/schema.ts and convex/_generated/ai/guidelines.md. While this establishes an ingestion point for untrusted data, this behavior is central to the skill's purpose as a code specialist.- [SAFE]: Instructions are consistent with the skill's stated purpose. Access is restricted to project-specific paths, and no patterns related to exfiltration, obfuscation, or privilege escalation were detected.

convex-expert