caveman
claudeTalk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Engineering harness: stop gate, git guard, format hook, session-start state recovery, /harness:init | audit | end.
Plugin harness from marketplace jiranon in Jiranon-K/claude-harness. 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.
auditAudit the harness in the current repository — score the five subsystems (instructions, state, verification, scope, lifecycle), check feature-list invariants, and report drift against the current harness templates. Use for "audit the harness", "is the harness healthy", "harness score", or after a har
endClose a working session cleanly — update the progress log and feature state with evidence, walk the clean-state checklist, write a handoff when the session was long, and commit when the repository is safe to resume. Use for "end session", "wrap up", "handoff", "before I stop", or when the Stop gate
initInstall or refresh the harness in the current repository — detect the stack, write harness.json, init.sh, feature_list.json, the progress log and docs/harness/, and add the harness block to CLAUDE.md. Use for "install the harness", "set up harness", "harness init", or when the Stop gate or SessionSt
principlesHarness-engineering principles for making coding agents reliable — five subsystems, repository as system of record, one active feature, evidence before done, deterministic gates over prompted rules, session handoff. Use when designing or changing CLAUDE.md, hooks, feature lists, progress logs, or wh
harness-creatorBuild, audit, and improve harnesses that make AI coding agents reliable: AGENTS.md/CLAUDE.md instruction files, feature/state tracking, verification gates, scope boundaries, session handoff, memory persistence, context budgets, tool-permission safety, and multi-agent coordination. Use this whenever
harness-evaluatorIndependent checker that scores a completed feature or session against the evaluator rubric using only evidence it can see (diff, test output, files). Use after a feature is marked passing, before accepting a session's work, or when the maker's own summary is the only proof of completion.
node "${CLAUDE_PLUGIN_ROOT}/hooks/session-start.mjs"
node "${CLAUDE_PLUGIN_ROOT}/hooks/git-guard.mjs"
node "${CLAUDE_PLUGIN_ROOT}/hooks/format-on-edit.mjs"
node "${CLAUDE_PLUGIN_ROOT}/hooks/stop-gate.mjs"
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.