diagram
Pass
Audited by Gen Agent Trust Hub on Jun 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill executes various utility binaries from the
~/.claude/skills/gstack/bin/directory to handle session state, telemetry, and project-specific settings. - [COMMAND_EXECUTION]: Utilizes
evalandsourceon the output of local scripts (gstack-slug,gstack-repo-mode) to dynamically configure the environment. - [COMMAND_EXECUTION]: Interacts with a local browser daemon to process Mermaid source code into SVG, PNG, and Excalidraw formats using a local HTML bundle.
- [EXTERNAL_DOWNLOADS]: Includes logic to check for software updates and send usage telemetry to the author's infrastructure, contingent on user approval.
- [EXTERNAL_DOWNLOADS]: References external documentation at
https://garryslist.org/posts/boil-the-oceanto provide context for its 'Boil the Ocean' design principle. - [COMMAND_EXECUTION]: Offers to modify the local project's
CLAUDE.mdfile to insert skill routing rules, which helps the agent identify when to invoke specific tools.
Audit Metadata