The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill establishes an automated auditing workflow that ingests and processes untrusted project source code, creating a potential vector for indirect instructions.
Ingestion points: The skills/accessibility-auditor.md file (Phases 1 and 2) directs the agent to utilize Glob, Grep, and Read tools to inspect **/*.swift files within the target project.
Boundary markers: The provided auditing instructions do not include specific boundary markers, delimiters, or warnings to ignore instructions embedded within the source code being analyzed.
Capability inventory: The skill utilizes file system exploration (Glob), content searching (Grep), and file reading (Read). It also coordinates with other specialized agents such as the accessibility-auditor and ux-flow-auditor.
Sanitization: There are no instructions for sanitizing, escaping, or validating the content retrieved from the project files before it is processed by the AI agent.

axiom-accessibility