Session discipline: verify claims, push back, stress-test plans with operator and adversary personas, and grade deliverables 0-100 with a reviewer agent.
Plugin rigor from marketplace andyhorn in andyhorn/claude-plugins. Commands read from the repo's manifest — 4 days 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.
grade-loopUse before declaring any deliverable done (a file you wrote or changed, a written plan, or a document). Dispatches the rigor:work-reviewer agent to run an acceptance gate first: every explicit requirement must be MET, every author claim must be VERIFIED, there must be no REFUTED claims, and there mu
plan-stress-testUse when planning a task of three or more steps, a plan that spends the user's money or hours, or a plan that changes shared or hard-to-reverse state, or immediately after such a plan is written. Runs two read-only persona agents in parallel, rigor:operator (the user's finite time, energy, interest,
adversaryPersona that tries to make a plan fail. Lists concrete failure modes, wrong assumptions, and missing dependencies, and checks stated facts against primary sources. Read-only. Dispatched by the rigor:plan-stress-test skill alongside the operator agent.
operatorPersona that represents the user and their finite resources (time, energy, interest, money). Reviews a plan and says what it costs the user, what they would drop, and what they would not finish. Read-only. Dispatched by the rigor:plan-stress-test skill alongside the adversary agent.
work-reviewerAcceptance-gate reviewer for a finished deliverable. Checks every explicit requirement and every author claim first; if the acceptance gate passes, scores correctness out of 75 and polish out of 25 for a combined 0-100 total. Tags each deduction as [BLOCKING] or [ADVISORY], caps the MUST-FIX list at
node "${CLAUDE_PLUGIN_ROOT}/scripts/session-rules.mjs"
Working-session workflows: align on a shared understanding before work starts, triage review findings into necessary, beneficial, defer, and discard, then walk the remaining decisions one at a time with docket, and keep a shared domain vocabulary with glossary.
4 skills
Dependabot PR triage: buckets open dependency PRs by provenance, semver, CI and mergeability, and analyzes release notes against your code. Read-only; never merges.
1 skill · 1 agent · hooks
Git workflows: rebase onto the default or stacked parent branch, driving mechanics on a cheap model and escalating conflict resolution to a strong one.
1 skill · 2 agents
Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 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
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.