bencium-harness
claudeDEPRECATED — renamed to bencium-loop. Install bencium-loop@bencium-marketplace instead; this alias will be removed.
6 skills · 9 commands · hooks
Lightweight AI-assisted build loop for Claude Code: scaffold + living roadmap + acceptance gate + mandatory plan/spec gate + verify (independent evaluator + gamed-test guard) + deploy/rollback. A memory that survives /clear. 9 commands, no infra.
Plugin bencium-loop from marketplace bencium-marketplace in bencium/bencium-marketplace. Commands read from the repo's manifest — 1 month ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
accelerated-modeSingle source of truth for the bencium-loop opt-in workflow fan-out — the two gates that must both hold, the fallback rule, the guardrails, and how a run is recorded in the marker. Referenced by every /bencium-* command that offers an accelerated path. Read this skill before invoking any Workflow.
context-loaderAuto-loads the bencium-loop project memory at the start of every session. Triggers when the cwd contains a .harness/ folder. Reads .harness/memory.md, .harness/rules.md, and the last 3 entries in .harness/archive/ to restore project context across /clear and session boundaries.
decision-watcherWatches for architectural decisions during a bencium-loop session and nudges the user to log them via /bencium-decide. Triggers when the conversation involves picking between technical alternatives, choosing a library/framework, defining a data model, setting a policy, or making any choice that futu
invariantsThe ordering and evidence rules the whole bencium-loop loop rests on, stated once with their reasons and their real enforcement status. Referenced by every /bencium-* command instead of being restated in each one. Read this before deciding whether a step may proceed.
markersThe shared contract for the machine-readable JSON marker blocks that bencium-loop commands emit at the end of a loop step — placement, content rules, and versioning. Referenced by every /bencium-* command that emits one; the per-command schemas stay in their commands.
phase-bannerSingle source of truth for the bencium-loop phase color scheme, banners, emoji, and progress strip. Referenced by every /bencium-* command so visual styling stays consistent across the loop. Read this skill before printing any phase header.
/bencium-loop:bencium-decideLog an architectural decision to .harness/archive/ and surface it in memory
/bencium-loop:bencium-deployVerify → deploy → health check → log. Refuses deploy if /bencium-verify fails.
/bencium-loop:bencium-featureAppend a new feature to tasks.md ## Roadmap and optionally promote to Now
/bencium-loop:bencium-initInitialize the build loop in this repo (greenfield or brownfield, auto-detected). 5-question interview, single LLM pass, ~60s.
/bencium-loop:bencium-nextPick the next task from tasks.md ## Now, write a plan/spec, gate on user approval, then build
/bencium-loop:bencium-promoteMove items between Roadmap→Now in tasks.md, or demote stale lines from memory.md to archive
/bencium-loop:bencium-retroPostmortem on a failure. Writes archive entry, proposes memory.md and ACCEPTANCE.md updates.
/bencium-loop:bencium-rollbackRun the configured rollback command and log the reason
/bencium-loop:bencium-verifyWalk ACCEPTANCE.md against actual repo state, report PASS/FAIL/SKIP with evidence
${CLAUDE_PLUGIN_ROOT}/hooks/plan-gate.sh
${CLAUDE_PLUGIN_ROOT}/hooks/acceptance-gate.sh
${CLAUDE_PLUGIN_ROOT}/hooks/deploy-gate.sh
${CLAUDE_PLUGIN_ROOT}/hooks/post-commit-nag.sh
DEPRECATED — renamed to bencium-loop. Install bencium-loop@bencium-marketplace instead; this alias will be removed.
6 skills · 9 commands · hooks
Detects user communication style (high-context relational vs low-context transactional) and adapts responses accordingly.
1 skill
Answer Engine Optimization for AI search visibility. Optimize content for ChatGPT, Claude, Gemini, AI Overviews citations.
1 skill
Code style, tech stack, and workflow conventions for React/Next.js/TypeScript, TailwindCSS, Supabase projects.
1 skill
Expert UI/UX design guidance for unique, accessible interfaces. Always-ask-first protocol for visual decisions, colors, typography, layouts.
1 skill
Production-grade frontend interfaces that avoid generic AI aesthetics. Based on Anthropic's Frontend Designer Skill with bold creative commitment.
1 skill
Talk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.
1 skill
Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications
5 skills · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Package any website or local web build into a lightweight desktop app with Pake CLI.
1 skill
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.