doc-driven
Warn
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: MEDIUMREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill relies on
npx @docmd/coreto build, validate, and serve project documentation. This command downloads and executes external code from the npm registry at runtime, which is a form of remote code execution from an unverified source. - [COMMAND_EXECUTION]: The instructions direct the agent to execute various shell commands to manage the documentation lifecycle, specifically
npx @docmd/core build,npx @docmd/core validate, andnpx @docmd/core devwithin thedocs/directory. - [EXTERNAL_DOWNLOADS]: The skill requires fetching the
@docmd/corepackage and its dependencies from the npm registry. It also suggests installing it as a local dev dependency ifnpxinvocation fails. - [INDIRECT_PROMPT_INJECTION]: The skill has a significant attack surface for indirect prompt injection because it is designed to read and process large amounts of untrusted data from the repository to generate documentation.
- Ingestion points: The skill reads the README, manifests, entry points, implementation source code, and project tests (as specified in
setup.md). - Boundary markers: No explicit delimiters, boundary markers, or specific "ignore embedded instructions" warnings are defined for the ingested file content.
- Capability inventory: The skill can execute shell commands via
npxand has write access to the filesystem to create and modify documentation files. - Sanitization: There are no instructions for sanitizing or validating the content read from the repository files before it is processed by the agent.
Audit Metadata