The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon data from external, potentially untrusted sources such as PR/MR comments and CI logs.
Ingestion points: The agent fetches review threads, CI failure logs, and SCM state via CLI tools and APIs.
Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following instructions that might be embedded within external comments or log files.
Capability inventory: The skill possesses the capability to modify code, run tests, push to remote repositories, and interact with SCM APIs (e.g., resolving review threads via GraphQL).
Sanitization: The instructions do not specify any filtering or sanitization of the external content before the agent processes it.
[PERSISTENCE]: The skill establishes a long-running execution pattern by using the ScheduleWakeup tool to re-trigger itself autonomously. While this is the core intended functionality, it allows the agent to maintain a persistent presence and continue operations without direct user oversight for extended periods.

loop