maniple

maniple is an MCP server (plus Claude Code slash-command workflow) that orchestrates multiple subordinate Claude Code or OpenAI Codex sessions using local terminal backends (tmux or iTerm2). It can spawn worker sessions, manage them with git worktrees/branches for isolation, send messages, read logs, and monitor lifecycle/events; it also supports an HTTP service mode with streamable HTTP transport.

Evaluated Mar 30, 2026 (0d ago)
Repo ↗ DevTools mcp claude-code codex orchestration tmux iterm2 git-worktrees multi-agent local-automation
⚙ Agent Friendliness
57
/ 100
Can an agent use this?
🔒 Security
28
/ 100
Is it safe for agents?
⚡ Reliability
30
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
75
Documentation
70
Error Messages
0
Auth Simplicity
95
Rate Limits
0

🔒 Security

TLS Enforcement
60
Auth Strength
15
Scope Granularity
0
Dep. Hygiene
45
Secret Handling
30

No authentication or authorization model is described for MCP/HTTP endpoints, so security appears to rely on local network/process boundaries. Terminal automation and log reading/writing imply sensitive data could end up in JSONL logs and worktrees unless users manage redaction/retention. HTTPS/TLS requirements for HTTP mode are not documented; score is therefore moderate-to-low. Dependency list includes iterm2 and mcp but no vulnerability/CVE posture is provided.

⚡ Reliability

Uptime/SLA
0
Version Stability
35
Breaking Changes
40
Error Recovery
45
AF Security Reliability

Best When

You want an agent-driven coordinator to launch and manage multiple local LLM coding sessions with visible terminal screens and per-worker git isolation.

Avoid When

You cannot safely use local terminal automation (tmux/iTerm2) or cannot tolerate background sessions/worktrees and log files being created on the host.

Use Cases

  • Running parallel Claude Code/Codex sessions for separate tasks (e.g., frontend vs backend)
  • Offloading implementation to background workers while keeping the coordinating agent’s context small
  • Monitoring and interacting with spawned workers (interrupt, take control, close)
  • Isolated experimentation using git worktrees per worker/issue
  • Adopting orphaned terminal sessions after MCP server restarts
  • Using HTTP mode for persistent MCP communication and worker polling

Not For

  • Environments without tmux/iTerm2 support (since it relies on terminal backends)
  • Teams needing remote/cloud execution isolation (it appears focused on local terminal session orchestration)
  • Use cases requiring formal auth/tenancy controls via the API
  • Highly security-sensitive workflows where terminal/log artifacts must never be written to disk

Interface

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

Authentication

OAuth: No Scopes: No

No explicit authentication mechanism is described for the MCP tools or HTTP mode. Access control is likely implicit (local execution / whatever Claude Code provides).

Pricing

Free tier: No
Requires CC: No

No pricing model described; appears to be a developer tool/library.

Agent Metadata

Pagination
cursor_none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • Operations that spawn workers/branches may be non-idempotent if repeated without checking existing sessions/worktrees.
  • Correct backend setup is required (tmux or iTerm2, plus iTerm2 Python API enabled).
  • In HTTP mode, ensure transport configuration is compatible with MCP client expectations (streamable-http).
  • Using worktrees creates persistent filesystem artifacts (.worktrees) and branches that may affect repo state if not cleaned up.

Alternatives

Full Evaluation Report

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

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