Open sourceRust · ≈94k stars · Apache-2.0
Impeccable
A design vocabulary for coding agents: polish, bolder, critique, audit.
Impeccable gives coding agents a design vocabulary. Commands such as polish and bolder steer the work, and critique and audit review it. A design hook scans each UI file the agent edits and catches AI slop before the agent calls the work done. Install it with npx impeccable install.
Pros & cons
Each point cites its sources and says what kind of evidence backs it. A second reviewer tried to refute every claim before it was kept.
Pros
- DocsShips a 62-rule deterministic slop detector that runs without an LLM, and `npx impeccable detect --github` emits GitHub Actions annotations for CI.[1]
- DocsOffers 24 named commands (e.g. `bolder`, `quieter`, `critique`, `audit`) so the agent follows a specific design playbook instead of vague prompts.[1]
- DocsOne `npx impeccable install` detects and configures 20+ harnesses, including Claude Code, Cursor, Codex CLI, Copilot, Gemini CLI and OpenCode.[1]
- MetadataApache-2.0, about 94.7k stars, 31 open issues and commits as of 2026-10-29.[1]
Cons
- User reportsThe skill runs a self-contained native engine binary downloaded to `~/.impeccable/bin/`, and a Windows build was flagged by ESET and Windows Defender as a trojan until a 4.0.3 update.[1][2]
- User reportsReleases have shipped broken installs, including a missing checksum file ("Could not verify skill bundle: HTTP 404") and a 4.2.1 build that read 0 of 35 reference files; both were later closed.[1][2]
- DocsSource-file scans apply pattern rules only; contrast and tiny-text rules need a rendered URL, so static CI runs miss part of the rule set.[1]
- DocsLive mode runs your package.json `impeccable:manual-edit-validate` script with your permissions, so it needs review in unfamiliar repos.[1]
- Best for
- Front-end teams that want an agent to iterate on existing UI with a repeatable design critique plus a deterministic anti-slop lint in CI.
- Watch out for
- It depends on a downloaded native binary, so check AV and supply-chain policy before rolling it out on managed machines.
Assessed Oct 29, 2026 · adversarially reviewedSuggest a correction
Neighbours on the workbench
MDX · ≈307k stars · MIT
obra/superpowers
An agentic skills framework and software methodology that works.
Maintained by Jesse Vincent (obra)
Shell · ≈306k stars · MIT
mattpocock/skills
Small, composable agent skills for real engineering work.
Maintained by Matt Pocock (AI Hero)
JavaScript · ≈292k stars · MIT
affaan-m/ECC
Agent harness performance optimization system: skills, instincts, memory and security for coding agents.
Maintained by Affaan Mustafa
Connections
Relations from the Harness Hub knowledge graph, as charted in the State of Harness reports.
Maintained by · 1
- pbakausOrganisation
Similar entries · 2
- Taste SkillProject
- UI/UX Pro Max SkillProject
Sources
Verified Oct 29, 2026