agentsh
agentsh is an agent shell for AI agents: execution-layer security that intercepts file, network, and process activity at runtime with deterministic policy and audit logs.
agentsh at a Glance
- Pricing
- Free
- How you use it
- CLI, shell shim, or harness wrapper
- Who it's for
- Teams running AI agents and harnesses
- Works with
- Claude Code OpenAI Codex CLI Cursor Aider E2B Daytona
agentsh Key Features
- Approval gates
- Command steering
- File and network interception
- Process activity auditing
- Runtime policy enforcement
Can't Find What You're Looking For?
We build custom AI agents tailored to your specific business needs. From ideation to deployment, TinyAI.Tools is your partner in AI innovation.
No commitment required. Let's discuss your ideas.
agentsh Alternatives
prismor
Prismor intercepts every tool call your AI agents make, enforces security policies, and logs a full audit trail. Works with all enterprise AI agents.
Long Horizon
Your coding agent writes the feature — let it test it too. Long Horizon runs real browser tests and produces shareable execution reports with logs, screenshots, and network detail for confident feature delivery.
Containarium
Open-source agent runtime with SSH-native isolation. Each agent gets a persistent, network-isolated box on Kubernetes or LXC — no kube-apiserver token, eBPF egress policy, MCP server built in. Self-hostable, Apache 2.0.
open-managed-agents
Open Managed Agents with Claude Managed Agents and OpenAI Agents API compatibility. Hosted or self-hosted, with your choice of models, harnesses and sandboxes.
metronix-memory
Metronix is open-source, self-hosted memory infrastructure for AI agents with persistent memory, structured state, permissions, checkpoints, and audit trails.
Clausura
CI-native agent CLI tool for deterministic pipeline gating. - liuyanghejerry/Clausura