html-testing-checklist
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources including bug trackers (Monday, Linear, Jira, GitHub, Asana), pull request diffs, and code commits to populate the checklist content.
- Ingestion points: External bug tracker records (titles, bodies, comments), PR diffs, and commit history are parsed and used to generate checklist steps.
- Boundary markers: The skill instructs the agent to HTML-escape all source-derived values and use
textContentwithcreateElementfor DOM construction, explicitly forbidding the use ofinnerHTMLfor untrusted data. - Capability inventory: The skill creates HTML files on the local filesystem and provides the ability to perform write operations back to issue trackers using existing tools (e.g., closing tickets, adding comments).
- Sanitization: Instructions mandate a redaction step that uses regex patterns to identify and replace credential-shaped values (such as API keys, tokens, and secrets) with placeholders before embedding them in the artifact.
Audit Metadata