release-openclaw-announcement
Pass
Audited by Gen Agent Trust Hub on Jul 23, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill incorporates an indirect prompt injection surface by processing external, potentially untrusted content from the web.\n
- Ingestion points: The skill reads release notes from GitHub, commit messages, and package registry metadata to draft announcements.\n
- Boundary markers: No specific delimiters or "ignore instructions" warnings are defined for the ingested external data.\n
- Capability inventory: The skill has the ability to post content to a Discord client using the
$discord-user-postcommand.\n - Sanitization: No explicit validation or sanitization steps are defined for the data retrieved from external sources.\n- [COMMAND_EXECUTION]: The skill generates shell commands for the announcement text, such as
openclaw update --channel beta --yes, which are intended for end-users to copy and run.\n- [EXTERNAL_DOWNLOADS]: The skill fetches configuration and release metadata from the project's official website (openclaw.ai) and GitHub. These are recognized as legitimate project resources.
Audit Metadata