AutoGen maintenance mode makes your framework exit plan urgent
Microsoft has put AutoGen into maintenance mode: its README says new features and enhancements are over, and the project is community-managed. That turns every new AutoGen dependency into an exit-path decision, while coding-agent evaluation is concentrating around platforms that keep expanding: Cursor leads outbound clicks this week after adding a headless CLI, TypeScript and Python SDK workflows, and MCP and ACP modes; GitHub Copilot is up 42.3% week-over-week. Separately, WebRun rose from 13 to 22/36 and is now classified as a Domain Specialist (L3). If AutoGen sits in your architecture, inventory its dependencies and require a migration path before you add another workflow.
See what changed on AutoGen's listing →
What moved today
|
22/36 · Domain Specialist
Score now 22/36 (13→22). The rubric moved on documented capability evidence; press alone does not move the score.
Trigger: submission.
|
Hyper
Re-scored up (20→22)
22/36 · Domain Specialist
Score now 22/36 (20→22). The rubric moved on documented capability evidence; press alone does not move the score.
Trigger: submission.
|
What was reported, but the rubric didn't move
|
16/36 · Adaptive Collaborator
OpenAI released ChatGPT Images 2.5 on September 8, with testing showing sharper visuals, richer textures, more natural lighting, and up to 50% lower latency than its predecessor. The launch comes as image generators compete to become dependable parts of professional creative workflows. For those tracking AI video and media generation, this signals a faster push toward image tools that can feed broader multi-asset production pipelines. The key trade-off is still unresolved: ChatGPT Images 2.5 leads on visual quality and speed, but Nano Banana 2 may retain an edge on structured prompt precision.
|
What's new in the directory
|
10/36 · Guided Assistant
Netcore is a cloud marketing platform for teams that manage customer engagement, personalization, and retention across web and mobile. It positions agentic workflows with business context and guardrails to help turn customer signals into campaigns and revenue outcomes across nine-plus channels.
Source: published listing, rubric-scored on ingest.
|
|
3/36 · Reactive Tool
SlopSpot Pro is a Chrome extension that flags overused ChatGPT-style phrasing in web pages and editable text fields. It runs locally, giving writers a quick way to spot repetitive AI-sounding language without sending text to a remote service.
Source: published listing, rubric-scored on ingest.
|
|
6/36 · Guided Assistant
iLook is a free web-based face analysis tool that evaluates one authorized adult photo for visible facial characteristics, including face shape, symmetry, a subjective face score, and Golden Ratio context. It also provides appearance and style-oriented feedback, with results retained in a private history until you delete them.
Source: published listing, rubric-scored on ingest.
|
What changed on the ground
Changes we detected on each tool's own pages — pricing, features, platform reach — via our daily crawls. Often ahead of any press coverage; each links to the scored listing.
|
14/36 · Adaptive Collaborator
Anaconda acquires Kilo Code — Kilo Code’s website now announces that the company has been acquired by Anaconda, linking to Anaconda’s acquisition announcement.
Source: the tool's own site — detected 2026-09-14 by agentic.ai's daily recrawl.
|
|
13/36 · Adaptive Collaborator
Microsoft puts AutoGen into maintenance mode and directs new users… — AutoGen’s README now states that the project is in maintenance mode, will receive no new features or enhancements, and is community-managed.
Source: the tool's own site — detected 2026-09-14 by agentic.ai's daily recrawl.
|
|
14/36 · Adaptive Collaborator
Asana acquires StackAI to build agentic work management workflows — StackAI’s website announces that Asana has acquired the company, positioning the deal around bringing human-agent workflows into a unified work-management platform.
Source: the tool's own site — detected 2026-09-14 by agentic.ai's daily recrawl.
|
|
8/36 · Guided Assistant
Mem0 expands agent integrations with Codex, OpenCode, OpenClaw, Hermes, n8n… — Mem0’s updated site documents broad platform expansion across agent runtimes and workflow tools, including new plugins, MCP/tool support, lifecycle hooks, persistent memory workflows, and integrations for OpenAI Codex, OpenCode, OpenClaw,…
Source: the tool's own site — detected 2026-09-13 by agentic.ai's daily recrawl.
|
|
10/36 · Guided Assistant
Pipedream repositions as an integration layer for AI agents with 10,000+ tools — Pipedream has substantially repositioned its product around Connect, offering managed authentication, MCP access, prebuilt tools and triggers across 3,000+ APIs, and a raw API proxy for AI agents.
Source: the tool's own site — detected 2026-09-13 by agentic.ai's daily recrawl.
|
|
18/36 · Adaptive Collaborator
Cursor expands its agent platform with a new CLI, SDK, ACP integrations… — Cursor’s updated site documents a substantially broader agent platform, including a headless CLI with interactive, plan, ask, shell, MCP, and ACP modes; TypeScript and Python SDK workflows for local and cloud agents; and third-party editor…
Source: the tool's own site — detected 2026-09-13 by agentic.ai's daily recrawl.
|
From Agentic-News.ai
Agentic-AI industry signal we're tracking — launches, releases, and research from across the ecosystem. The directory rubric hasn't scored these yet; awareness now, evaluation next pass.
|
EasySpecs.ai
product launch
EasySpecs.ai, built by Spaii Labs SL, opened public access to a platform that documents codebases and turns change requests into verified specifications. The launch comes as AI coding agents produce code faster than teams can review it, pushing quality checks earlier in the development process. For those tracking agentic AI product launches, this signals a growing market for tools that supervise and constrain coding agents before they take action. Its real test is whether spec verification catches defects that traditional code review misses.
Via Agentic-News.ai — TechBullion. Read: https://techbullion.com/easyspecs-ai-launches-spec-review-platform-so-teams-trust-ai-code-nobody-has-time-to-read
|
|
FINTRX
product launch
FINTRX launched Fin, an AI agent that brings private wealth intelligence into email, Slack, Microsoft Teams, Outlook, and Google Calendar. Fin continuously monitors the industry and proactively sends alerts, prospect insights, and other information. For those tracking agentic AI products, the launch highlights a shift from on-demand chat to agents embedded in daily work. The key question is whether Fin can turn those alerts into reliable next actions without overwhelming users.
Via Agentic-News.ai — Yahoo! Finance. Read: https://finance.yahoo.com/technology/ai/articles/fintrx-launches-ai-agent-named-130100857.html
|
|
Intermedia
product launch
Intermedia launched AI Receptionist on Sept. 10, 2026, bringing a native AI voice agent to Intermedia Unite and Contact Center. The product is built to capture inbound opportunities, reduce missed calls and streamline customer interactions. For those tracking agentic AI products, this signals voice agents moving from standalone demos into business communications suites. The unresolved question is how much of the follow-up workflow the receptionist can complete without human intervention.
Via Agentic-News.ai — Financial Post. Read: https://financialpost.com/globe-newswire/intermedia-launches-ai-receptionist-extending-its-digital-teammates-portfolio-with-native-ai-voice-automation
|
Biggest movers this week
Listings with the largest week-over-week jump in outbound-click interest
- GitHub Copilot · +42.3% week-over-week. GitHub Copilot helps you write, review, and adapt code directly in GitHub, your IDE, and the terminal. It supports everything from inline suggestions to agentic coding workflows with broader model choices and enterprise controls.
- OpenHands · +31.3% week-over-week. OpenHands (formerly OpenDevin) is a fully autonomous open-source AI coding agent. MIT-licensed core, plans + executes Python actions via CodeAct, uses browser + terminal + files. ~72% on SWE-bench Verified with Sonnet 4.5 via the OpenHands Software Agent SDK (arXiv:2511.03690).
- Codex CLI · +16% week-over-week. Codex CLI is OpenAI’s coding agent that runs on your computer from the terminal. It can also be used in VS Code, Cursor, Windsurf, or as a desktop app.
- Claude Code · +5.3% week-over-week. Claude Code is Anthropic's agentic coding tool that lives in your terminal. It understands your entire codebase, makes multi-file edits, runs commands, manages git workflows, and uses MCP for tool integration. Built with a Unix philosophy — it reads, plans, edits, and verifies in a loop. The fastest-growing product in the coding agent category.
Biggest declines this week
Listings with the largest week-over-week drop in outbound-click interest
- Aider · -51.7% week-over-week. Aider helps you edit code from the terminal with an AI model connected to your codebase. It runs locally in your repo and can work with models from providers like xAI using an API key.
- ElizaOS · -40% week-over-week. ElizaOS is an open-source platform for building, deploying, and managing AI agents. It includes a CLI, web UI, and plugin-based architecture so you can run agents locally and extend them with integrations and custom logic.
- Cline · -7.1% week-over-week. Cline is an AI coding assistant for VS Code that can inspect your project, edit files, run terminal commands, and use a browser while asking for permission at each step. It is aimed at developers working on real codebases who want more than code completion.
Agenticness Leaderboard
This week's most-clicked tools, re-ranked by rubric score. The score is the editorial signal; the popularity is what drew the directory's traffic here this week.
Claude Code is Anthropic's agentic coding tool that lives in your terminal. It understands your entire codebase, makes multi-file edits, runs commands, manages git workflows, and uses MCP for tool integration. Built with a Unix philosophy — it reads, plans, edits, and verifies in a loop. The fastest-growing product in the coding agent category.
GitHub Copilot helps you write, review, and adapt code directly in GitHub, your IDE, and the terminal. It supports everything from inline suggestions to agentic coding workflows with broader model choices and enterprise controls.
- Cline · 18/36 · Adaptive Collaborator · L2
Cline is an AI coding assistant for VS Code that can inspect your project, edit files, run terminal commands, and use a browser while asking for permission at each step. It is aimed at developers working on real codebases who want more than code completion.
- OpenHands · 18/36 · Adaptive Collaborator · L2
OpenHands (formerly OpenDevin) is a fully autonomous open-source AI coding agent. MIT-licensed core, plans + executes Python actions via CodeAct, uses browser + terminal + files. ~72% on SWE-bench Verified with Sonnet 4.5 via the OpenHands Software Agent SDK (arXiv:2511.03690).
What operators are searching for this week
A week's demand signal — what operators are typing into the directory's search. Suggestion-chip defaults + off-topic / greetings filtered.
- "best coding agent"
- "AI agent to triage and reply to my email inbox"
- "best coding agent for a large existing codebase"
- "tools to automate customer support ticket triage"
- "AI agent that can browse the web and fill out forms"
Learn: How Pragor actually works
Pragor is a coordination layer for AI agent teams, not the agent itself. It sits around frameworks like CrewAI, LangGraph, and AutoGen and gives humans and agents a shared project board. That board keeps durable context, so work does not disappear after one turn or one run. It also adds approval workflows, which means the agent can prepare work, but a person can still review and green-light actions before they go forward. The system keeps an attributable record of work, so you can see what happened and who did it. In plain English: Pragor helps organize agent work, preserve state, and route decisions through humans when needed. It does not claim to make the agents fully autonomous on its own.
See Pragor's full scorecard →
One tool worth testing
Pipedream — Pipedream is worth testing for teams that need an agent-facing integration layer rather than custom OAuth and connector plumbing. Its site now positions Connect around managed authentication, MCP access, prebuilt tools and triggers across 3,000+ APIs, plus a raw API proxy—an explicit shift toward agent use. In a 30-minute pilot, connect one agent to a production-adjacent SaaS workflow and compare its authentication, tool coverage, and failure handling with your existing integration path.
Our pick: ChatGPT Deep Research
The use case: research & deep analysis.
Tops our independent /36 — strongest on autonomy and planning, ahead of Gemini Deep Research. Pricing: Subscription (quotas undisclosed).
Ranked by our 9-dimension Agenticness rubric — not by who pays us (nobody does). The full scorecard, true-cost pricing, and the reliability evidence behind every score are in the report.
Read the full report →
|