greptile
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the
greptileCLI globally using the commandnpm i -g greptile. This downloads code from the public NPM registry at runtime. - [DATA_EXFILTRATION]: The core functionality involves using the
greptileCLI to perform branch reviews. This process sends repository code, branch diffs, and commit metadata to the external servicegreptile.comfor analysis. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes repository code through an external AI service, creating a surface for indirect prompt injection.
- Ingestion points: The skill reads the repository's feature branch and commit history during the
greptile reviewprocess as described inSKILL.md. - Boundary markers: There are no specific boundary markers or instructions to ignore embedded prompts within the processed code mentioned in the skill.
- Capability inventory: The skill executes shell commands via the
greptileCLI, performs network operations togreptile.com, and instructs the agent to analyze and act upon the findings generated from the external source. - Sanitization: No sanitization or validation of the code content or the resulting review findings is defined in the skill instructions.
Audit Metadata