The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Linear issue titles and descriptions, creating a surface for indirect prompt injection where malicious content in an issue could attempt to influence the agent's classification or actions.
Ingestion points: The skill fetches issue content using get_issue and query_data (SKILL.md, Step 1 and Step 4).
Boundary markers: The instructions do not specify the use of delimiters or clear "ignore embedded instructions" warnings when processing the fetched text.
Capability inventory: The skill has the capability to modify workspace data using the save_issue tool (SKILL.md, Step 6).
Sanitization: While no formal data sanitization is mentioned, the skill requires a mandatory human review and confirmation step (AskUserQuestion in Step 5) before any labels are applied, which serves as a primary security control.

linear-sdk-telemetry-labeler