Havuz

Micro-VM sandbox for AI agents: isolated, gated and auditable, to prove what your agents did

Architecture quality gate for polyglot codebases: declare components and allowed dependencies, fail the build on drift

AGENTS.md contract for AI coding agents + zero-dependency linter + SessionStart hook. Works with Claude Code, Codex, Cursor, Copilot.

Set up and manage self-hosted GitHub Actions runners on Linux and Windows, in WSL or Docker

CI-friendly CLI and API that fails builds when deployed URLs are missing or misconfigure HTTP security headers (HSTS, CSP, COOP, cookies...)

The cheapest way to run GitHub Actions: AWS spot machines in your own account at AWS's price, nothing added ($0.0008 a minute for 2 vCPU). O

git diff for CAD assemblies — two STEP files in, a picture of what changed plus a machine-readable change list out.

Catches multilingual voice agents placing the wrong order. τ-bench style benchmark + CI gate for voice ordering agents in English and Hingli

Turn a deno.lock diff into prioritised supply-chain signals — new raw-URL imports, integrity changes, registry moves.

CI regression gate for MCP servers: run a golden set of tool calls, diff against a baseline, fail the build when outputs get worse

✅ Acceptance is open-source e2e testing for Agents. Your coding agent turns a prompt into observable checks, runs the real product, and publ

Typed semantic decisions for Unix pipelines and CI, powered by TypeSafe AI Jev.

PantheonSim: GPU simulator for CUDA and ROCm. Run CUDA and HIP programs unmodified on a CPU, no GPU needed: simulated NVIDIA and AMD GPUs fo

Code review via the change impact measure

Static Ansible linting in Go: the 38 ansible-lint rules decidable from YAML alone, byte-for-byte pep8 output, no Ansible runtime.

Keep your sensitive credentials out of the agents you build and the coding agent you use. Test any agent with planted fake keys, or see whic

前端工程视觉与色彩全栈守卫：静态配色校验、渲染后视觉计量、证据驱动的 UI 评审闭环与规则演进治理。A full-lifecycle frontend visual guard: static contrast checks, post-render metrics, evide

Isolated install-and-smoke test drives for DeepSeek Harness plugins: installs a repo or npm package into a throwaway DSH_HOME profile, verif

MOVED to dsh-github-bot — https://github.com/PerryLink/dsh-github-bot. This repository is a pointer/backup; the old npm package @perrylink/d

A simple read-only proxy cache for git.

Your coding agent gaslights you. GODKILLER doesn't. Disk gates > vibes. claim_done only after verify.

Snapshot testing for LLM apps and agents, built to run locally and block regressions in CI.

Sign Windows executables with your Certum SimplySign cloud cert from Linux, macOS or Windows — pure HTTPS, no SimplySign Desktop.

End-to-end retail lakehouse on Databricks: dual-path ingestion from PostgreSQL and S3, dbt-modeled galaxy schema with SCD2 history, Airflow

Warm, disposable GitHub Actions runners with Docker Sandboxes microVM isolation across Linux, macOS, and Windows hosts. Designed for teams t

Empirical cost measurement and assertion tooling for Soroban — catch budget-exhaustion failures in CI before they fail on the network.

Tests CLI behavior from plain YAML: exit codes, output, files, snapshots, and interactive terminals (PTY/TUI)

OpenCode GitHub agent in GitHub Actions

Wireshark for MCP. A transparent proxy between your AI client and MCP server. Watch every call live in your terminal, fail CI on what it fin

Find gaps between .env, .env.example, shell variables, and Python code.

Proof-of-done for AI coding agents. Local evidence receipts for Codex, Claude Code, Cursor, and GitHub Actions.

Agent Reliability Analyzer for AWS CodePipeline and CodeCatalyst: find agent reliability issues

Validate Mermaid diagrams in Markdown! Catch real syntax errors using the official parser, built for CI.

Portable, always-fresh docs for agents and humans. A Surface hub is a conformant OKF concept; Surface adds the freshness OKF leaves out - it

Agent Reliability Analyzer for Azure Pipelines: find agent reliability issues

Open-source eBPF compatibility evidence and CI gate

Upkeep — an AI audit crew for your repo. Catches docs/spec/asset drift with evidence; output-only. Claude Code plugin/skill + reusable CI wo

Predict the silent integration traps an AI agent would ship (zero-revenue configs, auth bypasses, immutable wrong choices) — then enforce, i

Run coding agents unattended. Agents write the code. Axocoatl writes the receipt. Claude Code, Codex or a local model in a sandbox; the host

Nix binary cache and build matrix for GitHub Actions: eval once, build checks in parallel jobs, deduplicated uploads, automatic GC, zero set

boots your firmware and inference on N real targets. 0 setup, 0 hardware to own.

CI-native agent CLI tool for deterministic pipeline gating.

Reproducible artifact manifests, sandbox-output verification, and evidence-bundle validation for artifact-heavy workflows.

Reads your Terraform plan and Kubernetes changes and posts a plain-English risk summary on every pull request, blocking the dangerous ones

Automated quality checks that nudge AI coding agents toward better habits

Verification-gated skill routing and self-improvement harness for Hermes-style agent skills

On-demand, ephemeral Forgejo Actions runners on cloud VMs — kept warm only for the billing hour you've already paid for, so you never pay fo

Static analysis and health checks for dbt projects.

The cross-harness, self-improving memory layer for AI agents. Join our community: https://discord.com/invite/zdNChS2fBu

Deterministic semantic linter and PR reviewer for IEC 61131-3 Structured Text. 80 checks across diff-based, static integrity, FB-instance, P