blink-github
Pass
Audited by Gen Agent Trust Hub on Apr 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
blinkCLI tool to perform environment status checks, repository cloning, and authentication token retrieval. - [DATA_EXFILTRATION]: Facilitates access to repository content, issue metadata, and pull request information through the GitHub REST API.
- [PROMPT_INJECTION]: The skill processes content from GitHub (issues, PRs, files) which may contain malicious instructions designed to influence agent behavior (Indirect Prompt Injection).
- Ingestion points: External data is ingested via the
blink_connector_exectool when fetching repository contents, issue bodies, or PR details. - Boundary markers: The instructions do not specify any delimiters or safety prompts to isolate untrusted external content from the agent's core instructions.
- Capability inventory: The agent has the ability to create issues, open pull requests, and push code to GitHub repositories.
- Sanitization: There is no description of sanitization or validation protocols for data retrieved from GitHub before it enters the agent's context.
Audit Metadata