verification

Rigorous engineering workflows for writing less, higher-quality code and parallelizing with verifiable evidence.

Gitリポジトリで共通利用する最小限の開発ルールと、証拠に基づく検証・改善の進め方

Verified, citable numbers about the world right now, each with a public tick and SHA-256 fingerprint.

Skills for long-running agent work — write the completion contract, carry a goal to a verified finish, keep review…

A scoped production-engineering workflow for implementing, debugging, reviewing, refactoring, and release-preparing…

Verification fixture exposing separate secret words through a skill and an MCP tool.

Paved engineering commands for AI coding agents: repository context, governed workflows, explicit Tools, verification…

100 AI skills for systems and hardware engineers: requirements review, electrical and mechanical interfaces, power and…

Independently falsify or approve implementation plans before execution with ACCEPT, MODIFY, or REJECT verdicts per task…

Read, verify, and curate a receipt-backed Distill research corpus, including bounded active-session worker handoffs.

Verification-first agent operating system powered by OpenCode Zen and OpenCode Go via ModelHitch.

The BootLoops protocol layer: seven disciplines for trustworthy quantitative work with agents — acceptance gates,…

Proposal-only CollectiveOS host projection for constraint-governed planning, evidence synthesis, and state-change…

Meta-skills that sharpen your other skills: promote repeated code into reusable scripts, audit skill descriptions for…

pstack engineering workflows adapted for Codex, based on irg1008/cstack.

GSPC: read the live measurement board, verify Ed25519-signed measurement cards and capsules, and list the…

Verified coding workflows plus a public context-aware commit skill that plan, implement, test, review, and summarize…

Three-layer verification pipeline for AI output. Extracts claims, finds sources, and flags hallucination risks so…

Automatically chooses and coordinates GPT models for project tasks, with the right effort and safeguards for each job.

Workflow rules from CLAUDE.md turned into hooks and skills that actually fire

Three-layer verification pipeline for AI output. Extracts claims, finds sources, and flags hallucination risks so…

Evidence-routed planning, implementation, and review workflows for high-correctness software changes.

Task contracts, bounded planning and orchestration, surgical implementation, debugging, research, review, verification,…

Efficient AI execution through focused context, bounded delegation, and evidence-based verification.

A guild of specialist agent prompts, and an inspector that can stop the job. 22 skills covering review and…

Run this workflow from Codex. Reverse the built-in Claude Code limitations discovered in the source code leak. Installs…

Knowledge-first workflow for agents backed by Usable. Search team knowledge before implementing, retrieve complete…

Flux Point engineering harness: one Definition of Done, one Evidence discipline, two drivers. Loop mode runs a…

Portable wcode Software Intelligence skill for MCP-capable coding agents.

DreamerOS starts agents with shared rules, verifies intent, carries memory across sessions, checks claims, and routes…

Three-layer verification pipeline for AI output. Extracts claims, finds sources, and flags hallucination risks so…

Feedback-based control system for coding agents.

Git workflow, versioning, worktrees, Justfile recipes, TDD, test shrinking, and best practices

Enforce a claim-to-proof workflow: define the evidence, run the right verification, and distinguish verified results…

Adaptive office engineering runtime: routed execution, independent review, verification, and closeout through one…

Vendor-neutralなState Machine、Machine Floor、REFINE連動のProfile自動判定、Capability SkillsでCodex/Devin系Agentを制御するSuzumemo Agent…

Blocks narrow fixes: a Hypothesis Ledger with a mandatory zoom-out hypothesis before any bug fix, root-cause diagnosis,…

A quiet, risk-scaled engineering method layer for Codex.

Run one piece of work across several herdr agent sessions, one git worktree per unit.

Personal workflows for engineering, review, investigation, verification, and craft.

CRAFT+ workflow for turning task descriptions into verified production-ready prompts.

Luna-first orchestration with a local Rust control plane, durable runs, and an MCP Apps workbench.

Auditable context selection, exact recovery, and receipts for AI coding agents.

Set up, inspect and update Accord's project-local requirements, V&V and engineering documentation workflow.

Verified, citable numbers about the world right now, each with a public tick and SHA-256 fingerprint.

KOSIF Think Pro 3 — think-first orchestration plus measured expert studio: image analysis, image generation, lighting,…

Design verification strategies that unlock higher levels of autonomous software development.

Lean verified agentic engineering: plan, implement, verify, doc, doc-edit, wrap and status skills plus deterministic…

Project-scoped Guardian operations with Cognitive Evolution, verified meta-learning, cross-Guardian knowledge transfer,…

Lightweight, evidence-first software delivery for Codex. 中文优先的 AI 软件交付工作流。

Beau's rigorous agent workflow. beau-mode routes each task to a playbook and the principles; skills for understanding…

Evidence-based architecture review, code audits, verification, and authorized improvements.

A human-governed, resumable engineering workflow for coding agents.

Enforce a claim-to-proof workflow: define the evidence, run the right verification, and distinguish verified results…

Opinionated, technology-neutral principles for project delivery.

Trust over speed for coding agents — the router (AGENTS.md + WORKFLOW.md), 11 discipline skills, and the mechanical…

Proseweaving: house standards and verification for agent-authored human-facing prose.

nobrainer-tech-flow: portable agentic workflows from task to checked result.

Implement changes with reproducible verification, visual evidence, and draft pull requests.

Evidence-first Brain workflows for inspecting GitHub repositories, diagnosing failures, and preparing verified changes.

Review, fix and explain code changes with concrete verification evidence and plain-language limits.

Git-based specification-driven development with central workflow coordination, design, specification, planning, task…

Local first, policy bound computer control with verified browser and desktop routes.

IRIS five-step controlled workflow with evidence, error capture, checkpointing, recovery, audit, and verification.

Intent to implementation: ready, then go.

Record a demo, tutorial, or proof movie of software actually running — and catch the silent defects (frozen picture,…

Portable Rootloom Agent Skills for inspectable code changes and evidence-backed review with bounded project and…

Three-layer verification pipeline for AI output. Extracts claims, finds sources, and flags hallucination risks so…

Your coding assistant plans, sandboxed workers build, and nomArmy checks every change before it counts: the tests run…

Evidence-first coding agent. Verifies before presenting. Attacks its own output with adversarial multi-model review,…

Create, remotely build, test, verify, and explore mobile apps from a cloud coding workspace using the full Revyl CLI,…

Blackadder DLC for Cursor: idea → plan → architecture + design system → walking skeleton → verified tracer-bullet…

An expert Game Boy Advance ROM-hacking workbench: scaffolds a game repo, builds it in Docker, verifies it in a headless…

Evidence observatory: x402/MPP inspection, signed receipt checks and settlement attestations.

Official AI Agent Skills Suite for SEIHouse Productions.

Sinch Build skills and developer documentation for Codex: Conversation API (SMS, RCS, WhatsApp), Voice, Verification,…

Lightweight high-agency coding with impact calibration, adaptive model routing, targeted verification, and bounded…

Evidence-driven engineering workflows for Codex and ChatGPT.

Composable workflow SDK, local CI runtime, and multi-target compiler for software verification — define checks once,…

Evidence-bound roadmap delivery from idea capture and research through shaping, planning, execution proof,…

Business verification (KYB) and agent identity (KYA) for AI agents, over Trulioo's hosted MCP server. Reads the tool…

Workflow orchestration, verification, and contribution skills for AI coding agents — plan-first execution,…

TCPA consent verification for lead generation. Verify lead consent, pull machine-readable proof, and manage consent…

Agylite: deterministic routing, budgeted context, project state and local test-first verification over a curated bundle…

GodPrompt engineering workflows for task routing, TDD, debugging, verification, and progressive disclosure through MCP.

The BootLoops research skills: heavier machinery for specific tasks — proving with agents (adversary-first pipelines),…

Twenty-four named engineering principles (prove it works, fix root causes, subtract before you add, build the lever,…