The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze potentially untrusted external data, which creates a surface for indirect prompt injection attacks.
Ingestion points: As outlined in references/microservices_analyst_scope.md and references/service_inventory_and_dependency_analysis.md, the skill requests access to data from diverse external sources, including service catalogs, API gateway logs, APM traces, CI/CD history, and incident postmortems.
Boundary markers: The skill's instructions do not include explicit delimiters, tags, or specific guidance to the agent to treat the content of these external data sources as untrusted or to ignore any embedded natural language instructions.
Capability inventory: A review of the skill's references and workflows shows no evidence of dangerous capabilities such as automated shell execution, file system modification, or network exfiltration tools; the skill primarily produces analytical reports and summaries.
Sanitization: There are no instructions for the agent to sanitize, escape, or validate the content of the ingested technical data before incorporating it into its analysis or output summaries.

microservices-analyst