dependabot-sweep
Pass
Audited by Gen Agent Trust Hub on Jul 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes
gh(GitHub CLI) andgitto manage pull requests and repository branches. These actions are standard for the skill's stated purpose of PR automation. - [EXTERNAL_DOWNLOADS]: The skill executes package manager commands (
npm install,yarn install,pnpm install) which download dependencies from official registries. The updates are sourced from PRs generated by the official Dependabot application. - [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes untrusted data from the repository environment.
- Ingestion points: Metadata from pull requests (titles, branch names) and content from dependency files (
package.json) and CHANGELOGs. - Boundary markers: The skill does not implement specific delimiters or 'ignore' instructions when processing external PR data.
- Capability inventory: The agent can modify local files, execute shell commands, and interact with the GitHub API.
- Sanitization: The primary mitigation is the use of the
--author "app/dependabot"filter, which restricts the skill's input to PRs created by a trusted service.
Audit Metadata