docs-workflow
Warn
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: MEDIUMCREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill is designed to read and analyze project environment files such as
.envand.env.localto verify that documented variables are current. This involves accessing sensitive file paths that may contain secrets, although the skill includes instructions to exclude these secrets from final documentation. - [INDIRECT_PROMPT_INJECTION]: The skill processes content from various project files (e.g.,
README.md,CLAUDE.md, and source code) to generate audit reports and maintenance suggestions. This ingestion of untrusted local data creates a potential surface for indirect prompt injection. - [COMMAND_EXECUTION]: The audit process involves executing shell commands, specifically
git log, to compare documentation dates with actual file modification history. This is a standard part of the auditing workflow but involves local command execution. - [EXTERNAL_DOWNLOADS]: The skill performs network operations to check for updated versions of npm packages via
npm view. This interaction with the well-known npm registry is used to identify and flag outdated version references in the documentation.
Audit Metadata