The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its use of web-based market research.
Ingestion points: The 'Moat Analyst' sub-agent uses the WebFetch tool to retrieve data from external websites to research existing solutions and market evidence (SKILL.md).
Boundary markers: The skill does not employ specific delimiters or 'ignore embedded instructions' warnings when processing retrieved web content.
Capability inventory: The skill has access to powerful tools including Bash, Write, Edit, and agent orchestration capabilities (TeamCreate, Agent).
Sanitization: Content retrieved from the web is not sanitized or filtered before being synthesized into the final analysis.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to detect environment configuration (the CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS variable). This is a safe, read-only use of the shell for feature detection.

munger