The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection (Category 8) as it ingests untrusted data from repository files to drive autonomous actions.
Ingestion points: Processes repository content, CONTEXT.md, and Architecture Decision Records (ADRs) to inform the architecture loop.
Boundary markers: The skill lacks explicit boundary markers or instructions to disregard potential malicious directives embedded in the analyzed files.
Capability inventory: Possesses the ability to write files, create issues, and spawn sub-agents that modify the codebase.
Sanitization: No sanitization or validation of the content retrieved from the repository is performed before it is used to influence agent behavior.
[COMMAND_EXECUTION]: The skill identifies and executes 'repo-specific verification commands' and setup tools like setup-matt-pocock-skills. If a repository contains malicious configuration files (e.g., a hijacked package.json or Makefile), the agent may execute dangerous commands during its autonomous verification phase.

architecture-autopilot