loopbench-research
claudecodexRead and write papers with a deterministic argument checker. Separate install unit from loopbench-skills.
2 skills
Loopbench workflows, explicit CLI setup, and structured human input.
Plugin loopbench-skills from marketplace loopbench in hideh1231/loopbench. Commands read from the repo's manifest — 1 month ago.
This plugin runs code on your machine — it starts local MCP servers. The exact commands are listed below; read them before installing.
challenge-decisionUse when a coherent non-trivial Loopbench work unit reaches commit-sized readiness, and also when stakes are high, a proposed contract materially changes goal/scope/gates, results are surprising, an oracle is weak, or the user asks for an adversarial or rubber-duck review. Work in a fresh/read-only
comprehend-contractUse before accepting a Loopbench contract or deciding a human gate. Explain the purpose, choices, evidence, and limits first; preserve the same claims and viewpoint when changing display detail; then follow learning=off/light/deep. Light permits an explicit explanation-only confirmation. Deep asks o
delegate-implementationUse to carry a non-trivial Loopbench task from workspace selection and contract drafting to a running implementation without making the user remember the procedure. Reconcile the plan, explain the decision, apply learning=off/light/deep, ask only what changes the contract or an irreversible outcome
loopbench-retrospectiveUse only when the user asks to improve the Loopbench workflow or when task histories contain recurring or high-impact evidence worth codifying. Inspect accepted revision diffs, gate events, blockers, finish rejections, adversarial reviews, and source-backed notes. Propose at most two reversible chan
scan-decisionsUse when an existing Loopbench project needs a decision audit — extract past decisions, surface current direction-setting choices, separate automatic follow-ups, and render an evidence-backed report. Trigger for requests such as "what decisions did this project make?", "what still needs my judgment?
setup-loopbenchSet up or repair the standalone Loopbench CLI for the loopbench-skills plugin. Use only when the user explicitly asks to install, set up, update, or repair Loopbench, or when a requested Loopbench workflow cannot run because the CLI is missing or incompatible. Do not trigger merely because the plugi
work-taskUse for non-trivial coding, debugging, research implementation, analysis, experiment, or unfamiliar-domain tasks that must be carried through to an externally checked stopping condition with Loopbench. Trigger when a `.loopbench/` project exists, a task ID is mentioned, or the user asks to plan, exe
loopbench-inputruns: loopbench-input-server
Read and write papers with a deterministic argument checker. Separate install unit from loopbench-skills.
2 skills
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.