Plugin
harness-designing-plugin
Harness Designing Plugin helps design teams turn their scattered AI setup — Notion prompts, Figma comments, AGENTS.md rules, chat histories — into a five-layer design harness that compounds across sessions and rotations. Four skills walk you through the assembly: • /hd:learn — Q&A about the five-layer concept (Context Engineering, Skill Curation, Workflow Orchestration, Rubric Setting, Knowledge Compounding) • /hd:setup — detects existing harness artifacts, pre-analyzes all five layers in parallel (5× harness-auditor + rubric-recommender), then walks you through link / critique / scaffold / skip per layer • /hd:maintain — captures lessons and promotes them to team rules in AGENTS.md with SHA-256 plan-hash proof-of-consent • /hd:review — audits harness health in a 2-batch parallel dispatch (5 layer-specialist agents + 2–3 context agents); critiques harness artifacts against quality rubrics Ships 9 sub-agents and 14 starter rubrics (distilled from pbakaus/impeccable, Nielsen's 10 heuristics, Material 3, Fluent 2). Each rubric carries a Scope & Grounding section with personas, user stories, realistic scenarios, and anti-scenarios. Validated across a six-repo pilot matrix. This plug-in is the companion toolkit to Bill Guo's Substack series on design harnessing. Structural inspiration: EveryInc/compound-engineering-plugin.
- Analysed
- Last observed
What the code does
What does this mean?
These are capabilities our static analysis found in the latest published version. They describe what the code can do, not whether it is malicious: many tools need shell access or the network to do their job. Use them to decide what to check before installing.
Static analysis found none of the capabilities we track, such as shell commands, network requests or file writes, in the latest version.
Having a capability is not wrongdoing; many tools need these to do their job. It tells you what to check before you install.
This is public metadata, not a safety certification or complete assessment. Static analysis cannot establish what a tool does at runtime.
Risk
Sign in to see this tool's overall risk level and what drives it. Free.
Coverage
What does this mean?
How much of the tool we could inspect: listed only, downloaded, partly analysed, or fully analysed. Less coverage means fewer findings can be trusted to be complete.
Detailed report
What does this mean?
The full report lists every finding with the file it came from, and every change we observed between versions.
1 finding from automated static analysis of the latest version, with severities and explanations.
Using this tool across your team? We can assess your whole inventory.
Request an inventory assessment