The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted external data, such as logs and incident reports, without specific guardrails against embedded instructions.
Ingestion points: The skill explicitly instructs the agent to gather and analyze 'error logs' and 'incident data' in SKILL.md and references/documentation-templates.md.
Boundary markers: No specific delimiters or 'ignore embedded instructions' directives are defined for the process of analyzing these external data sources.
Capability inventory: The skill utilizes the debugging and codebase-recon skills, which typically provide the agent with capabilities to read files, inspect environments, and execute testing scripts.
Sanitization: There is no mention of input validation, sanitization, or filtering for the external data sources processed during the RCA investigation.

find-root-causes