The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to audit local code repositories, configuration files, and dependency manifests to inventory identities and access controls. This ingestion of untrusted local data creates a potential surface for indirect prompt injection if project files contain adversarial instructions intended to influence the agent's behavior during the audit.
Ingestion points: Processes local files using Read, Glob, and Grep tools as defined in SKILL.md and the instructions.
Boundary markers: The instructions do not define specific delimiters or "ignore" instructions for the ingested content, though they advise using safe targets and redacted evidence.
Capability inventory: The skill allows for Write, Edit, and WebFetch operations across all scripts.
Sanitization: No specific sanitization or filtering logic for external content is defined in the instructions.
[SAFE]: All external documentation references and resource links target the official algolia.com domain, which is a well-known and trusted service for this domain of expertise.

algolia-enterprise-rbac