The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources, creating a surface for indirect prompt injection. * Ingestion points: The skill fetches content from user-provided URLs and issue trackers (GitHub, Linear) to gather context for ticket creation (SKILL.md). * Boundary markers: There are no instructions to the agent to use delimiters or ignore instructions found within the fetched external data. * Capability inventory: The skill can write files to the local filesystem in the '.scratch/' directory and publish new issues to external trackers via network requests (SKILL.md). * Sanitization: No sanitization or validation of the ingested external content is described.- [EXTERNAL_DOWNLOADS]: The skill is designed to fetch data from remote locations, including issue trackers and arbitrary URLs provided as arguments, to populate its context.- [COMMAND_EXECUTION]: The instructions direct the agent to invoke other skills or commands, such as '/domain-modeling' and '/pstack-swarm', to resolve project artifacts and delegate tasks.

to-tickets