claudecodex

Research & Experimentation

Research and experimentation as distinct phases: investigate competing explanations, choose informative tests, and preserve evidence-linked lessons across product discovery, algorithms, skills, historical strategy, and investment theses.

3 on GitHubby Gautam Bhakarv1.1.0MITupdated 5 days agoproductivity
/

Plugin experimentation from marketplace gb-plugins-marketplace in gautam-achieveai/ClaudePlugins. Commands read from the repo's manifest — 4 days ago.

What it installs

Skills (3)

  • evidence-research

    Find, assess and synthesize evidence from web sources, papers, documents and supplied data. Use for literature reviews, current-state research, comparing conflicting claims, mapping an unfamiliar topic, or gathering evidence before an experiment round. Produces cited findings, gaps and actionable re

  • research-learning

    Consolidate research and experiment results into evidence-linked, conditional lessons. Use to learn from successful or failed trials, reconcile contradictory findings, update a research playbook, or check whether prior lessons transfer before another research round. Not a generic coding retrospectiv

  • research

    Investigate competing ideas and choose what to test next from experiments, historical evidence, or usage data. Use for hypothesis-driven research, interacting algorithm or compression designs, agent-skill improvement, product discovery, historical strategy, and long-term investment theses. Resume on

Subagents (6)

  • evidence-investigator

    Investigate a bounded evidence branch for a research coordinator. Typical triggers include finding current primary studies, examining contrary evidence and checking a specific source collection. See When to invoke for examples. Return evidence and leads, not an experiment execution or final study ve

  • evidence-reviewer

    Independently verify a decisive research claim, experiment plan or result. Typical triggers include conflicting sources, causal interpretations and conclusions that may drive the next experiment. See When to invoke for examples. Review evidence without rewriting the study or inventing objections.

  • experiment-designer

    Design a discriminating investigation after research and recall. Typical triggers include interacting algorithm ideas, product assumption tests, historical backtests and thought experiments. See When to invoke for examples. Produces a protocol; does not execute it or claim results.

  • experiment-runner

    Execute an assigned experiment protocol within its authorized scope. Typical triggers include local algorithm comparisons, scenario calculations and historical replay with fixed inputs. See When to invoke for examples. Do not redesign the protocol silently or treat a proposed run as an observed resu

  • learning-curator

    Preserve conditional learning and continuity after a research or experiment round. Typical triggers include contradictory results, revising a supported lesson and preparing the next research pass. See When to invoke for examples. Do not write personal agent memory or change global skills without aut

  • research-lead

    Coordinate a continuing research and experimentation study when the whole investigation is delegated. Typical triggers include competing compression designs, product discovery rounds and investment-thesis research. See When to invoke for examples. Do not dispatch another lead inside the same study.

View source on GitHub

More from gautam-achieveai/ClaudePlugins

All 11 plugins →

code-reviewer

claude

Code review toolkit with a general Code Reviewer agent for independent feature and merge-readiness review, built as a funnel: a cheap eligibility gate, deterministic review tiering, a shared context pack, a correctness lane and repo-history lane alongside specialized agents for duplicate detection, EUII leak scanning, exception handling, test coverage, design and code simplification, over-engineering and scope creep, architecture, performance, schema and wire compatibility, and feature-flag rollout. Findings are mechanically anchored to the diff, adversarially verified before posting, then synthesized into root causes, graded for severity and merge-blocking lane, and sequenced into an ordered fix plan, with each agent declaring the model-intelligence tier its work actually needs. Includes provider-aware review automation, structured comment publishing, and batch PR review orchestration.

13 skills · 2 commands · 38 agents

3

development

claude

Development methodology toolkit with Agile team delivery for large, long-horizon features. The agile-development skill coordinates thin end-to-end slices, durable progress state, early and continuous manual testing, observed-defect regression tests, behavioral tests for material acceptance contracts, final coverage review, verification, and final review. Includes Architect, Manual Tester, Developer, Critic, and blind-spot agents; design-first brainstorming; autonomous design; scenario-driven development as the default testing flow (manual testing from the first runnable slice, defect regressions, behavioral contract tests, then coverage-guided tests after all scenarios pass), with TDD as an explicit opt-in; subagent-driven execution; course-correction when fixes stack; compound-learning lessons read back at plan time; hooks that block git bypasses and untested completion; completion verification; code-review triage; work-item drafting; and provider-agnostic work-on and implement workflows.

15 skills · 2 commands · 5 agents · hooks

3

ado

claude

Azure DevOps integration for work item management, PR publishing, iterative PR tending, autonomous work item implementation, and autonomous backlog processing. Orchestrates development methodology skills via the development plugin.

8 skills · 6 commands · 3 agents

3

gh

claude

GitHub integration for issue and project management, PR publishing, iterative PR tending, issue-driven implementation, and autonomous backlog processing. Orchestrates development methodology skills via the development plugin.

8 skills · 6 commands · 3 agents

3

progress-tracking

claude

Continuous tracking for large work items and agent teams, with sourced events, durable state, progress and no-progress reports, worker nudges, recovery, and Markdown, Mermaid, and accessible HTML views.

7 skills · 3 agents

3

debugging

claude

Log-first debugging methodology using structured JSONL logs queried with DuckDB. Includes debugging workflows, logging enablement for codebases, logging compliance review with calibrated Trace-coverage recommendations for AI-assisted debugging, and systematic root-cause debugging.

3 skills · 2 agents · 1 MCP server

3

Similar plugins

caveman

claude

Talk like caveman. Cut filler. Keep technical details explicit.

22 skills · 1 command · 3 agents · hooks

JuliusBrussee/caveman 119,462

claude-mem

claudecodex

Persistent memory for Claude Code: the next session starts where the last one left off.

29 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

Taste Skill

claudecodex

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

Leonxlnx/taste-skill 114,391

career-ops

claudecodex

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

career-ops-hq/career-ops 82,253

mem0

claudecodex

Cross-session memory and token savings for coding agents.

6 skills · 1 agent · 1 MCP server · hooks

mem0ai/mem0 68,893

pake

claudecodex

Package any website or local web build into a lightweight desktop app with Pake CLI.

1 skill

tw93/Pake 63,463

Maintain Research & Experimentation?

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.

claude & codex plugins11 plugins