|
Blastpen — somewhere for the agent to be wrong
An OS-isolated sandbox runtime for coding agents. Any CLI agent, a git worktree of its own, confined by the operating system.
|
83 % |
|
Tazzie.ai — Symphonic Intelligence
Autonomous coding agent, 8 frontier models, semantic memory, Python sandbox. Chat, code review, and autonomous missions.
|
82 % |
|
devrandom • Accountable AI Agents
Open-source infrastructure for AI agents you actually control: every agent gets an ID you issue, every action leaves a tamper-proof record, and you can cut i...
|
81 % |
|
Axe — Built to cut. Not to chat.
|
81 % |
|
VisvoAI — The Python toolkit for building AI agents
One interface to any model, an agent loop done right, and a full coding agent built on both — proof it holds up. Open source, MIT licensed.
|
81 % |
|
Runvault — the secure agent shell
An agent with a real Linux shell — and the secrets locked out. Credentials stay behind a server-side proxy the sandbox never sees, and every action is logged.
|
81 % |
|
Evershell — Runtime for free agents
Your agent gets a shell. You keep the keys. A runtime for free agents.
|
81 % |
|
Fold
A terminal-first IDE for macOS, built for running AI coding agents in parallel.
|
81 % |
|
Slick Agent — the AI agent that learns from its own work
Slick is the self-improving AI agent: it writes skills from experience, improves them during use, remembers you across sessions, and runs anywhere from a $5...
|
80 % |
|
The Best Way to Sandbox and Supervise AI Coding Agents | Forkbench
The ultimate app to sandbox your coding agents, protect your API keys and supervise every agent at a glance. Vibe code at full speed, in total control.
|
80 % |
|
Termaxa — a gate for the shell commands AI coding agents run
The agent's prompt shows you the command; Termaxa shows you the consequence. It previews the blast radius, backs up first, blocks the dangerous ones, and kee...
|
80 % |
|
Tektonix — autonomous coding agent
A self-hosted autonomous coding agent, now with a Windows app. Give it a plain-English goal and it plans the work, writes the code, runs your project's real...
|
80 % |
|
ssh.bot — Controlled SSH Access for AI Agents
Give AI agents secure, controlled SSH access to your servers. Granular permissions, full audit logs, temporary sessions. Free during beta.
|
80 % |
|
Wedge · Sandbox for code you didn't write
Wedge runs untrusted installs, coding agents and third-party builds inside an OS sandbox that gets nothing it didn't ask for: no keys, no secret environment,...
|
80 % |
|
kolu — the best way to run terminals
kolu is a browser terminal workspace that spans machines: point it at a box over ssh and the whole canvas runs there — terminals, code, git, agents — backed...
|
80 % |
|
Founder OS Kit — Governance-as-code for your AI agent team
The exact constitution, approval-gate, budget-envelope and kill/scale system used to run a real one-person software company with a team of autonomous AI agents.
|
80 % |
|
Use your agent. Record and limit its tool calls — Writ
Connect Writ to Claude, Codex, Cursor, or OpenCode. Record visible tool calls, add a limit, or run a bounded task.
|
80 % |
|
Atom++ — the Atom is here
The hackable editor, reborn as an agentic, AI-native editor. It plans, edits, runs, and verifies its own work — on your key, with any provider, no middleman....
|
80 % |
|
llmsnitch — Watch what your AI coding agent actually did
Local-only execution tracing for Claude Code. Every tool call your agent makes — including the file edits its permission classifier never sees — recorded to...
|
80 % |
|
Codrift: run many AI coding agents at once
A desktop control surface for running many AI coding agents at once. Live terminals, isolated worktrees, one diff, one shared memory, driven from the keyboard.
|
80 % |
Detected social media networks