Agentic-Flow — Watchlist

MCP100 Assessment

Watchlist / Tracking

Why we’re tracking it

Follow-up value is specific to Agent harness orchestration with MCP swarm tools and library-safe APIs: Do in-process orchestration and MCP entrypoints keep authority and tool contracts aligned as the harness evolves?

Reviewed Oct 5, 2026 · Documentation review; complete setup workflow not independently tested

Best fit

Building self-learning agent orchestration and multi-agent harness workflows.

What we know

Agent meta-harness for routing, orchestration, memory, MCP coordination, hooks/background workers and opt-in repair/evolution workflows.

Maintenance / activity evidence

agentic-flow 2.1.3 published 2026-09-16T04:07:42.086Z

Tracking evidence observed Oct 5, 2026. These dated signals do not establish current maintenance. Reviewed enrichment confidence: medium.

Verification gaps / watch-outs

Installation and comprehensive security have not been independently audited.
Broad host/project authority through configured agents/hooks. Writes: Generate/edit config; Autonomous code repair; Hook-driven project changes.
Provider/tool-specific.
Sensitive access: Source code, Agent memory, Provider credentials

A broad orchestration framework whose effective risk is determined by enabled agents, hooks and command backends.

Full authority, access and evidence limits are available in Review context and Access & risk.

Popularity

· 294.1k package downloads / 30 days

Downloads observed Sep 25, 2026.

Do not overwrite existing reviewed stars/download metrics with nulls from this enrichment batch.

All limitations / Review context

Do in-process orchestration and MCP entrypoints keep authority and tool contracts aligned as the harness evolves?

The bounded review of the publisher's documented Agent harness orchestration with MCP swarm tools and library-safe APIs and release notices identified no disclosed material condition preventing an accurately scoped public tracking entry. The resource's operational permissions remain uncertainty; this assessment does not certify installation safety.

A broad orchestration framework whose effective risk is determined by enabled agents, hooks and command backends.
Broad host/project authority through configured agents/hooks.
Provider/tool-specific.
Project context, memories and trajectories may persist locally.
Reviewed write actions: Generate/edit config; Autonomous code repair; Hook-driven project changes
Sensitive access: Source code, Agent memory, Provider credentials
Installation, client compatibility and comprehensive security have not been independently audited. Authentication and access depend on the configured services.
Runtime behavior, dependencies and undisclosed vulnerabilities were not independently audited.
Documented tool authority and any remote data handling must be checked before installation or use.
Tool inventory, client matrix and permissions completeness were not requirements for this Radar tracking review.
Documentation mixes stable/alpha generations.
Autonomous repair and command hooks can modify code.
Some advanced Darwin/evolution capabilities are explicitly opt-in.

Original activity observations

agentic-flow 2.1.3 published 2026-09-16T04:07:42.086Z

Technical identity

npm/agentic-flow

Exact implementation source: https://www.npmjs.com/package/agentic-flow ↗
Reviewed resource type: Framework
Repository identity: ruvnet/agentic-flow
Package identity: agentic-flow

Evidence & Sources

Evidence basis: Recorded source evidence
Setup runtime verification: Not independently runtime-tested by MCP100 for this setup workflow.

Reviewed Oct 5, 2026 · Latest source observation Oct 7, 2026

Package registry ↗
agentic-flow
Recorded package identity, distribution or download evidence.
Includes references whose attribution is not established.
GitHub repository ↗
ruvnet/agentic-flow
Repository reference; implementation attribution is not established for every document.
Includes references whose attribution is not established.

Setup · Watchlist

Framework setup

Publisher instructions.

Source evidence observed Oct 14, 2026.

Additional setup details required — see the specific gaps below.

Warning

Not runtime tested, no third-party installation, network authorization, paid calls, or mutation executed.
Do not claim production service is available based on README alone.
Agent framework with optional MCP; `init` prepares project; do not present framework install as universal MCP registration
V1 omitted @alpha selector in documented MCP JSON and did not clarify initialization.
Official npm publisher package page README inspected for this exact npm identity; static research only. Some setup statements are conservatively summarized; no source code or live client execution checked. V2 review did not run any live client.
Initialize framework separately; do not treat top-level framework CLI as stdio service.
Avoid write, payment, token transfer, publishing, sending, booking and remote account changes as tests.
Replace credential, action and registration placeholders; follow the matching publisher/client prerequisites. No execution performed.

Install framework / project integration
Publisher instructions
Agent framework with optional MCP; `init` prepares project; do not present framework install as universal MCP registration | Publisher npm README revisited on 2026-10-14; exact identity retained.
Reviewed setup example · Publisher instructions
npx agentic-flow mcp start
Reviewed setup example · Publisher instructions
npx agentic-flow@alpha init
Reviewed setup example · Publisher instructions
npx agentic-flow@alpha mcp start
Configure / authenticate
MCP100 setup guide
Framework with optional MCP subcommand; initialize project using `npx agentic-flow init` before use, and launch MCP only using its documented mode.
Previously reviewed @alpha channel client example; separate from default channel: Publisher example adds client server "claude-flow" with command npx and args ["agentic-flow@alpha","mcp","start"].
Check connection
Suggested check
Check the framework MCP process starts and its tools appear in the chosen host, not just that init finished.
Remove / uninstall / revoke
MCP100 setup guide
Stop the application/SDK integration, remove its project package/configuration using the original installer, and remove any separately configured MCP connector. Not a universal standalone MCP server.

Official Documentation & Sources

What it does

Agent meta-harness for routing, orchestration, memory, MCP coordination, hooks/background workers and opt-in repair/evolution workflows.

Capability: AI & Reasoning
Resource type: Framework
Best fit: Building self-learning agent orchestration and multi-agent harness workflows.

Representative uses

Initialize agent projects
Run MCP orchestration
Route tasks to agents
Capture learning/repair data

Limitations

Documentation mixes stable/alpha generations.
Autonomous repair and command hooks can modify code.
Some advanced Darwin/evolution capabilities are explicitly opt-in.

Reviewed access

Access & risk

Review the exact access boundary before connecting this MCP to accounts, credentials, or sensitive data.

Authentication: Provider/tool-specific.
Permissions: Broad host/project authority through configured agents/hooks.
Data handling: Project context, memories and trajectories may persist locally.
Cost: free
Risk and cautions: A broad orchestration framework whose effective risk is determined by enabled agents, hooks and command backends.

Access scope and cautions

Write actions: Generate/edit config Autonomous code repair Hook-driven project changes
Sensitive access: Source code Agent memory Provider credentials
Risk level: high

FAQ

Based on reviewed evidence as of 2026-10-07. Watchlist tracking is not a recommendation or a security certification.

Which tasks are documented for this implementation?

The reviewed tasks include: Initialize agent projects; Run MCP orchestration; Route tasks to agents; Capture learning/repair data.

What setup and authentication are documented?

Recorded installation/entrypoint step: npm install agentic-flow@alpha; npx agentic-flow mcp start. Recorded transport: stdio. Prerequisites: Node.js >=18. These are the documented setup details, not a claim that installation or client compatibility was independently tested. Authentication: Provider/model auth is host-defined.

Can it modify data or perform write operations?

Documented write actions include: Generate/edit config; Autonomous code repair; Hook-driven project changes. Recorded permission boundary: Broad host/project authority through configured agents/hooks.

What data-handling boundary should I consider?

Project context, memories and trajectories may persist locally. Recorded sensitive-access context: Source code; Agent memory; Provider credentials.

What limitations and risks should I check before use?

A broad orchestration framework whose effective risk is determined by enabled agents, hooks and command backends. Documentation mixes stable/alpha generations. Autonomous repair and command hooks can modify code. Some advanced Darwin/evolution capabilities are explicitly opt-in. Runtime behavior, dependencies and undisclosed vulnerabilities were not independently audited. Documented tool authority and any remote data handling must be checked before installation or use.

Is this an official MCP?

RUV / ruvnet is listed as First-party MCP in MCP100’s current evidence.