The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is instructed to ingest and process external source code which may contain malicious instructions designed to influence the agent.
Ingestion points: The Discovery and Deep Analysis steps in SKILL.md direct the agent to read and explore target module paths and individual source files.
Boundary markers: The instructions do not define explicit boundary markers or directives to isolate external code content from the agent's logic, although 'Pattern 2' (Explain, Don't Copy) provides a partial functional constraint.
Capability inventory: The skill is granted access to powerful tools including Bash, Task, Read, Write, Edit, Glob, and Grep as defined in the allowed-tools section of the SKILL.md frontmatter.
Sanitization: The skill's workflow lacks explicit sanitization, validation, or filtering of the external code content before it is processed by the agent.

code-analyzer