The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted information from external work requests and project management records, which creates a potential attack surface for indirect prompt injection.
Ingestion points: External request content, Linear issue descriptions, and user comments are processed during the starting point identification and context reconciliation phases (SKILL.md Step 1.2 and references/domain-context.md).
Boundary markers: The instructions do not define specific delimiters or instructions for the agent to treat ingested content as untrusted data separate from its primary instructions.
Capability inventory: The agent has permissions to create and update project records in Linear and to propose modifications to documentation within the repository.
Sanitization: There are no explicit requirements for the agent to sanitize or escape external content before incorporating it into drafts or published records.
[NO_CODE]: The skill consists entirely of instructional markdown files and templates; no executable scripts or external dependencies were included in the skill definition.

triage