The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes rule files from the repository and external content via WebFetch. This creates a surface for indirect prompt injection, where malicious instructions embedded in a rule file or a fetched website could attempt to override the agent's behavior.
Ingestion points: The skill uses Read and Grep/Glob to load rule files (e.g., rules/typescript/ts-readonly.md) into the agent's context. WebFetch is used to read external documentation from angular.dev and other community sites.
Boundary markers: The instructions define a strict "Review Process" and "Output Format" that provide structural boundaries, though there are no explicit delimiters or "ignore embedded instructions" warnings for the ingested content.
Capability inventory: The skill utilizes WebSearch, WebFetch, Read, and Grep/Glob to perform its tasks.
Sanitization: No explicit data sanitization or filtering is performed on the ingested content; the skill relies on the agent's logic to differentiate between the data being reviewed and its own instructions.

angular-best-practices-rules-reviewer