Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Auto-updated: 2026.3.3 additional fixes (auth labels, Discord mega-wave, exec heartbeat routing, compaction continuity, session startup date grounding)
What This Skill Does
Generates a pre-ship risk report for OpenClaw pull requests by diffing the current branch against main and dynamically exploring changed modules with read-only shell commands. Each finding is scored by severity and backed by evidence from the diff and codebase exploration. No approve/reject decision is made — the report gives information for the developer to act on.
Each finding includes both diff evidence and dynamic exploration output, so risk assessments are grounded in the actual codebase state rather than static checklists.
When to use it
- Checking blast radius before merging a core module change
- Catching version-specific gotchas after an OpenClaw upstream release
- Spotting contribution policy violations before submitting a PR
- Assessing cross-module coupling risk in a refactor
- Identifying missing test coverage for changed files
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Run a pre-ship report on my current branch
- 1Load STABLE-PRINCIPLES.md, ARCHITECTURE-MAP.md, EXPLORATION-PLAYBOOK.md, and VISION-GUIDELINES.md from the references/ directory, plus CURRENT-CONTEXT.md if present
- 2Run git branch --show-current and git diff --name-only main...HEAD to identify changed files
- 3Classify each changed file by module and look up its risk tier in ARCHITECTURE-MAP.md
- 4Run grep, find, and git commands from EXPLORATION-PLAYBOOK.md to measure consumer count and blast radius per module
- 5Score each finding individually on a 1-10 scale and produce the structured markdown report
Markdown report with a module risk summary table, severity-scored findings each containing diff evidence and exploration evidence, and an executive summary with top actions before publishing