The Agent Skills Directory

[PROMPT_INJECTION]: Indirect Prompt Injection Risk. The skill facilitates reading and processing tickets and issues from Linear Multiworkspace, which constitutes an untrusted data ingestion surface. Maliciously crafted content within a ticket could attempt to influence agent behavior.
Ingestion points: apideck.issueTracking.tickets.list and related methods in SKILL.md used to fetch ticket data.
Boundary markers: None explicitly implemented in the provided documentation to delimit external ticket content from system instructions.
Capability inventory: The skill allows network operations (via the Apideck API) to read, create, and update tickets. It does not perform subprocess execution or local file system writes.
Sanitization: No specific content sanitization or validation of the external ticket data is described in the integration examples.
[SAFE]: External Communications. The skill communicates with unify.apideck.com, specs.apideck.com, and linear.app. These are recognized as legitimate vendor and service provider domains.
[SAFE]: Credential Management. The skill correctly instructs the use of environment variables (process.env.APIDECK_API_KEY) for secret management and describes the use of a managed vault for downstream service credentials.

linear-multiworkspace