linear
Pass
Audited by Gen Agent Trust Hub on Mar 7, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill has a vulnerability surface for indirect prompt injection because it processes data from Linear issues (such as titles and descriptions) that could be controlled by external users.
- Ingestion points: Data is ingested from the
https://api.linear.app/graphqlendpoint, specifically from issue objects. - Boundary markers: There are no explicit instructions or delimiters provided in the examples to prevent the agent from obeying commands embedded within the retrieved issue content.
- Capability inventory: The skill has the capability to perform network requests via
curland interact with local Git repository states. - Sanitization: No sanitization or escaping of the retrieved API data is performed beyond basic validation of title length.
- [COMMAND_EXECUTION]: The skill uses shell commands and pipe chains to automate API interactions.
- Examples include using
curlwithxargsto process Git commit logs and update Linear issues dynamically.
Audit Metadata