Linear Agents (AIG, Signals, Interaction)

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly instructs capturing an agent_oauth_token and shows example code embedding actorToken and agentToken directly into HTTP headers (Authorization / Linear-Actor-Token), which requires the LLM/agent to handle and output secret values verbatim.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). SKILL.md describes Linear agent interaction as receiving webhook interaction events for issues (user-assigned/mentioned), where the webhook body includes issue.id and actor and the agent then posts/creates signals; the interaction/event payload is outsider-authored free text (e.g., issue content/mention context) entering via the webhook workflow without any requirement to preselect a specific item.