claude

agent-rigor

Six skills for working with coding agents seriously: methodical (spec-first execution, with a failure model for stateful work), review (independent second opinion on the plan before building, ideally from a different model family), verify (adversarial check of agent work against its spec), independent-tests (failure-model tests written from the spec alone), handoff (survive context exhaustion), reflect (proactive offers without nagging).

0 on GitHubby Chris Cromptonv0.3.0MITupdated 30 days agoworkflow
/

Plugin agent-rigor from marketplace agent-rigor in ccromp/agent-rigor. Commands read from the repo's manifest — 29 days ago.

What it installs

Skills (6)

  • handoff

    Write a handoff document so work survives a context reset. Use when a task is outgrowing a single context window, when you're pausing work someone (or a fresh session) will pick up later, or when the user asks to "hand this off" or "create a handoff document". A deliberate handoff beats lossy automa

  • independent-tests

    Have the tests for risky, stateful work written by an agent that has seen only the spec and never the implementation, so the tests encode what the invariants require rather than what the code happens to do. Use when a spec has a failure model (states, concurrency, restarts, retries, cancels), or whe

  • methodical

    Lightweight methodical execution for multi-step coding tasks. Spec first, task list, verify-as-you-go loop, strict escalation policy. Use when the user says "be methodical" / "work methodically" / "use a task list", or when starting any multi-step implementation task. A middle tier between raw promp

  • reflect

    Before ending an idle turn, re-read the conversation with fresh eyes and either make one or two specific, low-effort offers of help, or say nothing. Use at a natural pause when the user has gone quiet and work is parked. The discipline is proactive help WITHOUT nagging: silence is the default, and a

  • review

    Get an independent second opinion on a plan, diagnosis, or about-to-be-sent instruction, from a reviewer that judges direction and premises rather than spec conformance. Use before acting on a risky or wide-blast-radius decision, before dispatching work to someone else, or any time you catch yoursel

  • verify

    Independently verify that work an agent (or you) just produced actually satisfies its spec, before it reaches the human. Spawns a read-only verifier subagent that reviews the diff against the spec and returns a structured pass/fail verdict with evidence. Use before claiming a task is done, before op

View source on GitHub

Similar plugins

mattpocock-skills

claude

Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).

39 skills

mattpocock/skills 323,108

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

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

mem0

claudecodex

Cross-session memory and token savings for coding agents.

6 skills · 1 agent · 1 MCP server · hooks

mem0ai/mem0 68,893

context7

claudecodex

Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.

1 MCP server

upstash/context7 64,709

agent-browser

claude

Automates browser interactions for web testing, form filling, screenshots, and data extraction

12 skills

vercel-labs/agent-browser 50,373

Maintain agent-rigor?

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 pluginslisted