The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using npx for self-verification of code. These include npx tsc --noEmit to check for compilation errors and npx convex dev --once to verify the deployment state against the Convex backend.
[INDIRECT_PROMPT_INJECTION]: The agent is directed to ingest content from convex/_generated/ai/guidelines.md if it exists. As this file is processed as instructions for the agent's workflow and could potentially be modified by other processes or external inputs, it serves as a surface for indirect prompt injection.
Ingestion points: convex/_generated/ai/guidelines.md (identified in SKILL.md).
Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within this file.
Capability inventory: The skill has the ability to execute shell commands via npx and read/write files within the convex/ directory.
Sanitization: No sanitization or validation of the file content is specified before processing.

convex-expert