release
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local scripts and command-line utilities (
git,gh,npm,node) as part of a release workflow. These actions are aligned with the stated purpose of automating project releases. - [INDIRECT_PROMPT_INJECTION]: The skill relies on the output of local test commands and git status checks to drive the release flow. While this represents an ingestion surface for data, the risk is negligible as the data comes from the project's own codebase and standard tools.
- Ingestion points: Command outputs from
git,node, andnpm(e.g.,git branch --show-current, test result tables). - Boundary markers: None explicitly defined in the prompt instructions, though the procedural nature of the skill limits the potential impact of adversarial content.
- Capability inventory: File writes (
e2e/live/skip-streak.json), network operations (git push,gh workflow run), and shell command execution. - Sanitization: Not applicable as the skill consumes predictable output from its own internal scripts and standard utilities.
Audit Metadata