linear-core-workflow-a

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data from CSV files and reads existing issue content (descriptions, comments) from the Linear API to perform updates or transitions.
  • Ingestion points: Data enters the agent context via fs.readFileSync("issues.csv") in the bulk creation example and via client.issues() or issue.comments() when fetching data from Linear.
  • Boundary markers: No explicit delimiters or boundary instructions are used when handling external text data.
  • Capability inventory: The skill utilizes client.createIssue, client.updateIssue, and client.createComment within the Linear SDK to manage SaaS resources.
  • Sanitization: The skill relies on standard CSV parsing and Linear's API schema validation for data integrity.
  • Context: This surface is inherent to the skill's primary function of issue management and does not expose sensitive system capabilities.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 08:05 PM
Security Audit — agent-trust-hub — linear-core-workflow-a