The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze existing codebases to guide design decisions. This creates a surface for indirect prompt injection if the files being analyzed contain malicious instructions targeted at the AI agent.
Ingestion points: Phase 1 (Explore) instructs the agent to use tools like rg and read local manifests to understand the system.
Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore instructions embedded in the analyzed code.
Capability inventory: The skill utilizes file system read/write operations and command execution for verification (typecheck, lint, tests).
Sanitization: No specific sanitization or validation of the ingested code content is mentioned.

software-design