Verification before completion

Keeps a TSV decision log for long or unattended agent runs, one row per decision with what, why, evidence and result, so a reviewer can check the work later.

cursor/plugins

Verifies a delivery end to end by driving the real product on a CLI, web, desktop or iOS Simulator surface, capturing evidence and publishing a round with the lh CLI.

lobehub/lobehub

Runs an unattended iterate-until-verified loop in which a user-set verify command, not the agent's own claim, decides when the task is finished.

tanweai/pua

Keeps an agent focused on the requested work by applying a five-step ladder that checks for direct solutions, real gaps and speculative defenses before adding anything.

lennney/stop-that-shit

A skill your agent uses when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any…

farm-fe/farm

Adds short, pointed workplace reminders drawn from two Chinese essays on corporate culture, nudging the agent to prove results with evidence instead of polished reports.

tanweai/pua

Proves a Codewhale change in the real product: a stamped release build, an atomic local install, fresh-shell verification and manual QA that automated gates cannot cover.

codewhale-hq/Codewhale

Explains why editing CUTLASS fused-MHA headers in ONNX Runtime can leave stale CUDA kernels after an incremental build, and how to force and verify a real rebuild.

microsoft/onnxruntime

Records an annotated screen recording of the agent testing an app hands-on, then posts the video and a results summary to the PR and tracker issue.

michaelshimeles/skills

Read and write email through the Atomic Mail from an AI agent.

Atomic-Mail/atomic-mail-agentic

Self-check your branch before creating a PR — catch dead code, prevent new model-specific Python examples, verify accuracy/perf claims, validate PR title format, and confirm merge readiness.

vllm-project/vllm-omni

The binding directive for ultrawork mode: evidence-first delivery, a light or heavy process tier, and constant use of memory during the work.

code-yeongyu/lazycodex

A skill your agent uses when the user asks for an AWS architecture diagram — VPC/networking, event-driven, landing zone, multi-AZ, serverless pipeline, or any diagram built with AWS service icons.

sparklabx/drawio-ai-kit

Delivers a change in thin vertical slices, each implemented, tested, verified and committed before the next, using vertical, contract-first or risk-first slicing.

addyosmani/agent-skills

Pushes an agent to keep verifying and changing approach after repeated failures, using a diagnosis line, evidence-based completion and confirmation before risky edits.

tanweai/pua

Executes a written implementation plan task by task in the current session, with a progress ledger, test-first gates and one fresh-context review at the end.

jnMetaCode/superpowers-zh

A skill your agent uses when the user asks for an Azure architecture diagram — VNet/networking, App Service, AKS, landing zone, multi-region, or any diagram built with Azure service icons.

sparklabx/drawio-ai-kit

Confirms that a change to PlotJuggler 4 really works in the running app by proving the rebuild, launching with real data and measuring the result.

PlotJuggler/PlotJuggler

Runs the Antigravity CLI headlessly with the agy command to analyze a repository or carry out a coding task, then checks its JSON result and the diff.

CherryHQ/cherry-studio

SkillsBench task authoring — walk a contributor from idea to submission-ready task following CONTRIBUTING.md and the task-implementation rubric.

benchflow-ai/benchflow

Researches code with evidence: traces callers, imports and cross-repo links, diagnoses failures and reports findings with exact file and line references and a confidence label.

bgauryy/octocode

Review pull requests and local branches for vllm-project/vllm-omni with a frozen snapshot, module-design ownership, feature-design overlays, targeted validation, and concise evidence-backed findings.

vllm-project/vllm-omni

A skill your agent uses when an AI agent wants to participate in the Haidian Centennial Jing-Zhang AI Innovation Belt open call, follow changing materials and community discussion, generate or…

open-city-ai/haidian

A skill your agent uses when about to claim a result, effect, significance, or that an analysis reproduces, before reporting or writing it up - requires running the analysis fresh and reading the…

K-Dense-AI/science-superpowers

A skill your agent uses when the user asks for a BPMN diagram, swimlane diagram, business process map, or workflow diagram with roles/lanes and phases.

sparklabx/drawio-ai-kit

Decomposes a complex feature into tasks, dispatches parallel specialist agents with durable state, and supervises verification, QA review and retries.

first-fluke/oh-my-agent

Adds a second, stronger model that the main agent consults before major decisions, when stuck and before finishing, controlled by /advisor commands.

cursor/plugins

Checks generated or edited documentation against the source code, flagging invented symbols, outdated samples and unverifiable claims before publishing.

amElnagdy/guard-skills

A gate before merge: verify runs the real app against the spec, and review has a different model do a senior code review, without editing code.

jsmastery-pro/skills

Coordinates and recovers multi-stage Light research projects from a single passport file, with checkpoints, stale-work tracking and rerouting only when you approve.

Light0305/Light-skills

Runs an evidence-bound audit, daily or weekly review through the requirement-ledger CLI, binding scope and authority first and blocking the final report until a handoff check passes.

adand-91/gpt-6-astra-skill

A skill your agent uses when the user asks for a Databricks lakehouse architecture diagram — medallion architecture (Bronze/Silver/Gold), Delta Lake, Unity Catalog, workspace deployment…

sparklabx/drawio-ai-kit

Turns a one-line idea into a task brief of up to 4000 characters that an autonomous agent can run through /goal, with measured baselines and cheat-resistant acceptance checks.

KKKKhazix/khazix-skills

Pushes an agent that keeps failing or gives up to exhaust every option, using harsh corporate-pressure wording, a diagnosis line and a proactivity checklist.

tanweai/pua

Applies a 12-stage verified workflow, from research to deploy, to non-trivial coding tasks, scaled to lightweight, standard or full mode by task size.

artemiimillier/bulletproof

Proves the Puppetmaster Jev transition behaves as specified: unset never uses the network, opt-in may skip the conflict auditor, and later versions only observe.

professorpalmer/Puppetmaster

Final checklist before handing work back in the herdr-auto-title repo: review the diff, run make check, apply the comment and AGENTS.md rules, then report.

kryptamine/herdr-auto-title

Run the infrastructure health check and fix anything that fails

diet103/claude-code-infrastructure-showcase

Read-only detector that compares SPEC.md with the code and reports invariant, interface and task drift grouped by severity, without changing anything.

JuliusBrussee/cavekit

A skill your agent uses when the user asks for a GCP or Google Cloud architecture diagram — VPC/networking, GKE, Cloud Run, landing zone, multi-region, or any diagram built with GCP service icons.

sparklabx/drawio-ai-kit

Scores an agent's finished work with a three-stage pipeline: free mechanical checks, an advisory semantic review, and an optional multi-model consensus vote.

Q00/ouroboros

Human handoff, Linear branch names, reviewable (non-draft) PRs, the cursor GitHub label, Claude, Greptile, and Codex review comments, preview test steps, proof of work posted on the GitHub PR, and…

langfuse/langfuse

A work-discipline protocol that makes Opus 4.8 (or any non-frontier model) operate at Fable-5-grade quality.

cozytab/fable5-mode

A skill your agent uses when the user wants to set up / scaffold / install a file-based multi-agent orchestration system in a folder.

netwaif/multi-agent-starter

Applies Procoder's senior-developer discipline in a repository: run the commit gate, format through the binary and work through specs, plans and todos.

azrtydxb/procoder

A required checklist for after code changes: validate each changed layer, check that docs and other layers stay in sync, and test before committing or opening a PR.

ZeroDeng01/sublinkPro

Runs a math modeling contest pipeline for CUMCM and MCM/ICM entries in Codex CLI, with git checkpoints, verified solver code and human review at each stage.

RealSeaberry/AutoMCM-Pro

Pushes an agent that keeps failing, gives up or claims unverified success into a diagnosis, evidence and verification loop, with a Pi extension for persistent mode.

tanweai/pua