add-release-notes
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- Command Execution: The skill utilizes
pnpmto run local repository scripts, such asgenerate-changelog-markdown-filesandgenerate-component-changelog-entries. These are standard tools used to automate documentation updates within a development environment. - Indirect Prompt Injection Surface: The skill processes external content from files like
packages/components/CHANGELOG.mdand instruction sets in.bob/rules/(Ingestion points). The skill lacks specific delimiters or boundary markers for this data (Boundary markers). The agent is granted the capability to execute shell commands viapnpm(Capability inventory). No explicit sanitization or filtering of the changelog content is mentioned (Sanitization). While this represents a potential surface for indirect prompt injection, the skill includes a manual checkpoint requiring user review before committing changes, which acts as an important safety control. - External References: The skill provides a template for linking to documentation hosted on Vercel (
hds-website-X-Y-Z.vercel.app). Vercel is a well-known platform for hosting web documentation, and this reference aligns with the skill's stated purpose of updating documentation links.
Audit Metadata