linear-add-task
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied task details and is instructed to explore external codebases or documentation mentioned in those details to populate agent notes. This creates a surface where malicious instructions in the user's brief or external files could influence the agent's behavior during the task creation process.
- Ingestion points: The
task detailsargument and external files/docs named in the brief as mentioned inSKILL.md(Step 3). - Boundary markers: Absent. The instructions direct the agent to map the brief's content directly into the task title, description, and checklist.
- Capability inventory: Tools for listing and saving issues/projects (
list_projects,list_issues,save_issue), along with the agent's internal capabilities for codebase and documentation exploration. - Sanitization: No sanitization, escaping, or validation logic for the input brief or the external documentation content is defined.
Audit Metadata