The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from GitHub pull requests and issues, which could contain malicious instructions designed to influence the agent's behavior. * Ingestion points: External content from pull requests and issues referenced in SKILL.md. * Boundary markers: The instructions do not define specific delimiters (e.g., XML tags or triple backticks) to separate the untrusted PR/issue content from the agent's primary instructions. * Capability inventory: The skill does not request access to dangerous tools, file system writes, or network operations in its frontmatter. * Sanitization: There is no mention of sanitizing or filtering the external content before processing.

larp-detector