The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing external data to generate agent configurations.
Ingestion points: Reads all files in .vibeRig/requirements/ and retrieves task descriptions from Linear issues.
Boundary markers: No delimiters or instructions are specified to help the agent distinguish between project data and potential embedded instructions.
Capability inventory: Executes agent-creator to define new agents, updates the project configuration file (project.yaml), and removes existing agent files (.toml) upon user confirmation.
Sanitization: There is no evidence of filtering or escaping of the ingested data before it is used to formulate instructions for new agents.- [COMMAND_EXECUTION]: Uses standard shell utilities for local filesystem analysis and integrity checks.
Evidence: Employs find and ls for directory traversal and grep for validating the schema of generated agent files.

update-team