poco-claw

Poco is a self-hostable AI agent execution platform/web UI that runs agent tasks in an isolated container, supports project/task management and artifact viewing, and integrates agent tooling via MCP/Skills plus an embedded browser and IM integrations.

Evaluated Mar 30, 2026 (0d ago)
Homepage ↗ Repo ↗ Ai Ml ai-ml agent sandbox mcp skills web-ui self-hosted docker typescript python fastapi nextjs
⚙ Agent Friendliness
31
/ 100
Can an agent use this?
🔒 Security
28
/ 100
Is it safe for agents?
⚡ Reliability
15
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
30
Documentation
35
Error Messages
0
Auth Simplicity
60
Rate Limits
0

🔒 Security

TLS Enforcement
60
Auth Strength
20
Scope Granularity
0
Dep. Hygiene
30
Secret Handling
30

From provided materials: claims of containerized/sandboxed execution and self-hosting. However, no concrete details are provided about auth, scope granularity, TLS configuration, secret handling, or dependency/security posture. Treat as unverified beyond the stated sandboxing concept.

⚡ Reliability

Uptime/SLA
0
Version Stability
40
Breaking Changes
0
Error Recovery
20
AF Security Reliability

Best When

You want a local or Docker-deployed “coworker” style agent workflow with a UI and sandboxed execution, and you plan to integrate via MCP/Skills rather than a traditional REST API client.

Avoid When

You need a stable, documented REST/GraphQL/OpenAPI interface for reliable automation, or you require clearly specified auth/rate-limit/error-code behavior from a public API.

Use Cases

  • Autonomous web research and task execution in a sandboxed environment
  • Managing multi-step agent workflows with plan/queue/termination
  • Uploading and processing files for agent-driven analysis or document generation
  • Viewing and replaying agent artifacts and command/browser I/O sessions
  • Integrating agent interactions through IM providers (DingTalk/Feishu/Telegram)
  • Extending agent capabilities via MCP/Skills tooling

Not For

  • Using as a low-effort managed SaaS without any self-hosting/ops considerations
  • Handling workloads that require strong, explicitly documented enterprise security/compliance guarantees (not evidenced in provided materials)
  • Scenarios needing a fully documented public developer API/SDK for programmatic integration (no such contracts shown)

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
No
SDK
No
Webhooks
No

Authentication

OAuth: No Scopes: No

No authentication mechanisms, auth flows, or scope model are described in the provided README/snippets.

Pricing

Free tier: No
Requires CC: No

README mentions a 'cloud subscription' as coming soon, but no pricing/tier details are provided.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • Sandboxed container execution can affect performance and persistence (state/files) depending on how sessions/containers are managed.
  • Agent tool-calling behavior can vary by imported Skills/MCP tooling; lack of documented tool contracts may require iterative testing.
  • Embedded browser/autonomous execution can produce long-running actions; without explicit documented limits/timeouts, automation may be harder to control.

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for poco-claw.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-03-30.

6533
Packages Evaluated
19870
Need Evaluation
586
Need Re-evaluation
Community Powered