MCP.Market

unlimited-mcp server

by triumsebas·io.github.triumsebas/unlimited-mcp·v0.9.4·2 stars

Delegate coding and sysops tasks to cheaper agents or local LLMs via durable background job queues.

B78/100grade B
What users say
No reviews yet
Be the first
Safety scan
B78/100

full report

Adoption
New

2 stars17 downloads/wk

Reviews

Write one

Nobody has reviewed unlimited-mcp yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

unlimited-mcp tools (38, 13 write)

write = sends, deletes, buys or posts

Read from the package source without running it. The installed server may list more.

  • add_agent

    Add or replace an agent entry in config.yaml.

  • add_allowed_root

    Add a filesystem path to allowed_roots.

  • add_deny_path

    Add a path to deny_paths (always blocked, even if inside an allowed root).

  • add_host

    Add or replace an SSH host in config.yaml.

  • add_provider

    Add or replace a provider entry in config.yaml.

  • add_queue

    Add or replace a queue entry in config.yaml.

  • answer_worker_questionswrite action

    Write answers for a clarification round, unblocking the waiting agent.

  • await_job

    Block until job_id finishes and return its final result.

  • await_worker_questions

    Block server-side until a clarify job needs you, then return once.

  • cancel_jobwrite action

    Send SIGTERM to a running job and mark it cancelled.

  • cleanup_brancheswrite action

    Remove leftover unlimited-mcp/* branches from a git repository.

  • cleanup_jobs

    Evict old job directories from disk.

  • cleanup_remote

    Clean up accumulated state on a remote SSH host.

  • cleanup_state

    Clean up accumulated state: old log lines, /tmp files, and orphaned worktree dirs.

  • configure_agentwrite action

    Update params or top-level fields on an existing agent.

  • configure_safetywrite action

    Update global safety and clarify settings in config.yaml.

  • delegate_to_agent

    Delegate a coding task to a configured agent (aider, opencode, smolagents, ...).

  • detect_conflicts

    Report coding jobs whose changed files overlap — run before merging branches.

  • get_job_result

    Return the full result for a background job, with zombie detection.

  • get_job_status

    Return the current status of a background job (lightweight poll).

  • get_worker_questions

    Return all clarification rounds for a job and their answered status.

  • install_and_restart

    Install a Python package with uv/pip then restart the server.

  • list_capabilities

    List all configured agents, providers, tools, and the safety policy.

  • list_jobs

    Return jobs matching the filters — defaults to the inbox view.

  • list_safety_policy

    Return the current safety configuration, allowed_roots and deny_paths.

  • lookup_agent_cli

    Look up a CLI agent in the knowledge catalog.

  • query_logs

    Query the MCP operational and error logs without needing to know their paths.

  • register_agent_knowledgewrite action

    Write a new CLI entry to knowledge.local.yaml (gitignored user catalog).

  • remove_allowed_rootwrite action

    Remove a filesystem path from allowed_roots.

  • remove_deny_pathwrite action

    Remove a path from deny_paths.

  • remove_entrywrite action

    Remove a named entry from a config section.

  • restart_server

    Restart the MCP server process (re-exec with the same argv).

  • resume_agent_task

    Relaunch a failed clarify-phase job with the Q&A history injected.

  • run_and_summarizewrite action

    Run a command, wait for completion, then summarise stdout via the provider.

  • run_commandwrite action

    Run a command after the safety pipeline.

  • run_shellwrite action

    Run an arbitrary shell script via bash or sh.

  • ssh_trust_host

    Add the SSH host key to ~/.ssh/known_hosts via ssh-keyscan (one-time setup).

  • submit_taskwrite action

    Submit a command or agent invocation as an explicit background job.

Public scan report

scanner v0.1.9 · 2026-09-27 · same rubric, same numbers if you re-run it

no findings
  • Code scan55 source files scanned25/25
  • –Live reliabilityno gateway calls yet and no remote to proben/a
  • –Tool poisoningtools not inspected (local package is not executed); not countedn/a
  • Auth qualitylocal package, no credentials required12/15
  • Maintenancelast push 128 days ago8/15
  • Maintainer identityregistry namespace matches repository owner6/10
Overall 78/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

How unlimited-mcp compares

Against every graded MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score
78/100
higher than 57% of 48,255 servers
Tools
38
the typical server has 7
Tools that change things
34%
typical: 0%. Fewer write tools means less an agent can break
GitHub stars
2
more than 75% of 28,688 servers
Downloads a week
17
more than 5% of 12,191 servers with a package

What its tools let an agent do: 17 read · 7 create · 5 change · 4 delete · 5 run or deploy

Install unlimited-mcp in Claude Code, Cursor or VS Code

claude mcp add unlimited-mcp -- uvx unlimited-mcp
Add to Cursor

unlimited-mcp: common questions

Is unlimited-mcp server safe?
Mostly: it is graded B (78/100). Read the unlimited-mcp safety report
How do I install unlimited-mcp?
It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does unlimited-mcp need an API key?
Not as far as the registry entry and our scan can tell: no credentials are declared or required.
Is unlimited-mcp maintained?
The last commit was 165 days ago (2026-05-23). The latest release is v0.9.4.
How does unlimited-mcp compare with other MCP servers?
safety score 78/100 (higher than 57% of 48,255 servers); tools 38 (the typical server has 7); tools that change things 34% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.