desloppify

Pass

Audited by Gen Agent Trust Hub on Sep 24, 2026

Risk Level: SAFEPERSISTENCECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill makes extensive use of the desloppify CLI and system-level tools such as git, bun, pip, go, and gem to perform scans, manage worktrees, and install dependencies.
  • [PERSISTENCE]: The desloppify install-hooks command scaffolds a .githooks/ directory and configures the repository via core.hooksPath to ensure scanning logic executes automatically during the git lifecycle.
  • [DYNAMIC_EXECUTION]: The tool supports a plugin architecture that loads and executes JavaScript/TypeScript code from repository-local configuration files (e.g., desloppify.config.js) using standard module loading, allowing for custom auditing logic.
  • [EXTERNAL_DOWNLOADS]: The skill recommends installing various third-party tools (e.g., ast-grep, ruff, knip) via package managers and includes a benchmark fetch command that retrieves data from external manifests.
  • [INDIRECT_PROMPT_INJECTION]: The agent triages and acts upon findings generated from the scanned repository's source code, which could potentially contain patterns designed to mislead the agent's decisions.
  • Ingestion points: Artifacts stored in .desloppify/reports/ (e.g., latest.findings.json, latest.report.md).
  • Boundary markers: The instructions do not specify explicit delimiters or isolation instructions when the agent reads findings.
  • Capability inventory: The agent can perform code modifications via desloppify fix and manipulate git worktrees.
  • Sanitization: There are no documented sanitization steps for findings before they are processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 24, 2026, 07:51 AM