linear
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from Slack threads to perform Linear operations, creating a potential surface for indirect instructions.
- Ingestion points: Data is ingested from Slack conversation history and thread context as described in SKILL.md.
- Boundary markers: There are no explicit delimiters or specific 'ignore instructions' markers used when the agent reads external Slack data before drafting Linear issues.
- Capability inventory: The skill utilizes MCP tools to create and update issues, add comments, and change workflow states, which are sensitive write operations.
- Sanitization: The skill mandates stripping Slack-specific identifiers, @mentions, and conversational noise, and requires generalizing transcripts, which serves as a defensive measure against malicious payload injection into the final ticket.
Audit Metadata