The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes logs and codebases, which are external data sources that could contain malicious instructions designed to exploit the agent's logic.
Ingestion points: The skill reads external data using tools like Read, Grep, Glob, and Bash to search logs and codebases (SKILL.md).
Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore or isolate instructions that might be embedded within the files being analyzed.
Capability inventory: The skill is configured with powerful tools including Bash, Write, Edit, and MultiEdit, which increases the potential impact if an indirect injection were successful (SKILL.md metadata).
Sanitization: The instructions do not include steps for sanitizing or validating the content extracted from logs before the agent processes or acts upon it.

error-detective