Claude Flow (claude-flow) MCP Server
@claude-flow/mcp
npm
v3.0.0-alpha.9
Published by @claude-flow — no publish provenance and no public repository, so the publisher could not be verified and the source cannot be independently located.
Standalone MCP (Model Context Protocol) server - stdio/http/websocket transports, connection pooling, tool registry
The grade answers one question — how safe is this server for you to adopt — so it is computed in two auditable stages. Nothing below is an opinion or an LLM's guess; every line is a real term the deterministic engine applied, and the same input always yields the same number.
1. Threat score — 100 − 6.3 = 94. What the published surface and source actually contain:
| Points | What was found | Category |
|---|---|---|
| −6.3 | Untrusted input concatenated into a command sink MTC-SRC-009 | injection |
2. Client adoption risk — 94 − 8 = 86. Three small, subtract-only factors that reflect your risk in adopting it — a clean scan proves less on a powerful, unverified or barely-inspectable package, so the grade says so plainly:
| Points | Adoption-risk factor |
|---|---|
| −6 | capability blast radius (high) — client exposure if the model is manipulated |
| −2 | publisher verification (unlinked) — no provenance/repo link, but the shipped source was fully read |
Capability observations and info notes are shown under Findings but never scored.
Open any row's finding below for the file, line and evidence behind a deduction.
Findings 9
high Shell/command execution in server code (.claude/helpers/github-safe.js)MTC-SRC-002
In the server's implementation (`.claude/helpers/github-safe.js:12`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: port { execSync } from 'child_process'; import { writeFileSync, unlinkSync } from 'fs'; import { tmpdir } from 'os'; imp
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server .claude/helpers/github-safe.js
high Shell/command execution in server code (.claude/helpers/learning-service.mjs)MTC-SRC-002
In the server's implementation (`.claude/helpers/learning-service.mjs:24`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: execSync, spawn } from 'child_process'; import Database from 'better-sqlite3'; const __filename = fileURLToPath(import.
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server .claude/helpers/learning-service.mjs
high Shell/command execution in server code (.claude/helpers/metrics-db.mjs)MTC-SRC-002
In the server's implementation (`.claude/helpers/metrics-db.mjs:12`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: port { execSync } from 'child_process'; const __dirname = dirname(fileURLToPath(import.meta.url)); const PROJECT_ROOT =
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server .claude/helpers/metrics-db.mjs
high Shell/command execution in server code (.claude/helpers/statusline.cjs)MTC-SRC-002
In the server's implementation (`.claude/helpers/statusline.cjs:28`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: { execSync } = require('child_process'); // Configuration const CONFIG = { enabled: true, showProgress: true, sho
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server .claude/helpers/statusline.cjs
high Shell/command execution in server code (.claude/helpers/statusline.js)MTC-SRC-002
In the server's implementation (`.claude/helpers/statusline.js:11`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: { execSync } = require('child_process'); // Configuration const CONFIG = { enabled: true, showProgress: true, sho
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server .claude/helpers/statusline.js
medium Hardcoded egress to an external endpoint (dist/sampling.js)MTC-SRC-003
In the server's implementation (`dist/sampling.js:231`): A hardcoded outbound call to a fixed external host inside server code is a classic exfiltration/telemetry channel — especially paired with reads of local data. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: const response = await fetch('https://api.anthropic.com/v1/messages', { method: 'POST',
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server dist/sampling.js
medium Hardcoded egress to an external endpoint (src/sampling.ts)MTC-SRC-003
In the server's implementation (`src/sampling.ts:324`): A hardcoded outbound call to a fixed external host inside server code is a classic exfiltration/telemetry channel — especially paired with reads of local data. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: const response = await fetch('https://api.anthropic.com/v1/messages', { method: 'POST', headers: {
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server src/sampling.ts
medium Untrusted input concatenated into a command sink (.claude/helpers/github-safe.js)MTC-SRC-009
In the server's implementation (`.claude/helpers/github-safe.js:101`): A shell/process command assembled from concatenated or interpolated values is command injection when any part is attacker-influenced — the OWASP canonical RCE flow. Verify what reaches the interpolated value. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: t, execute normally execSync(`gh ${args.join(' ')}`, { stdio: 'inherit' }); } } else { // Other commands, execut
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server .claude/helpers/github-safe.js
medium Dynamic module load from a non-literal (.claude/helpers/learning-service.mjs)MTC-SRC-005
In the server's implementation (`.claude/helpers/learning-service.mjs:467`): Loading a module chosen at runtime (from a variable) can pull in and run attacker-influenced code paths. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: mizedEmbedder } = await import(agenticFlowPath); this.embedder = getOptimizedEmbedder({ modelId: 'all-
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server .claude/helpers/learning-service.mjs
What this scan could not see
- No tools were enumerated, so prompt-injection, capability and toxic-flow analysis had no tool surface to inspect. To grade a package’s real runtime tools, scan the running server: --command "npx -y <package>".
Versions 1
Scan history per published version. The engine is deterministic — the same version always yields the same score, so a changed score means the package itself changed.
| Version | Score | Findings | Engine | Scanned |
|---|---|---|---|---|
v3.0.0-alpha.9 latest |
B 86/100 | 9 | 1.13.0 | 2026-08-25 |
Embed this score
Show this server's live Trust Score in your README, docs or website. The badge is served straight from the registry and updates automatically after every rescan — no API key needed. It links back to this page, so anyone who sees the grade can also read the findings behind it instead of taking a number on faith.
Verify this score yourself
The score above is reproducible: the same package version always yields the same result. Run it locally or over the free API — no account, no LLM, fully deterministic.
npx mcptrustchecker scan @claude-flow/mcp --onlineOther implementations of Claude Flow 1
Independent packages implementing the same tool, scanned with the same engine. Compare all 2 side by side →
More in Data Science & ML
MCP server for 2s.io — 575+ pay-per-call tools for AI agents — ground-truth data, AI gateway, and agent infra (storage, locks, queues, watchers). x402 USDC on Base/Solana, no API keys, upto usage billing, free trials.
ArXiv preprints + Google Scholar papers, with citation counts in one query.
Add MCP servers to your favorite coding agents with a single command.
Help agents automatically write and test stories for your UI components
Model Context Protocol server for AI-Archive platform - enables AI agents to discover, submit, and review research papers
MCP server for Semantic Scholar research workflows with stdio and Streamable HTTP transports.