Cursor expands CLI, SDK, and MCP support
Today’s site changes all point in the same direction: these tools are moving from standalone chat into operating surfaces. Cursor added CLI installation and interactive-session docs, expanded SDK streaming and interaction updates, and broadened MCP and cloud-agent configuration. Claude now stretches across Cowork, Code, Design, Science, and Microsoft 365, Chrome, Slack, and Google Workspace connectors, while ChatGPT rewrote plan features around GPT-5.6, scheduled tasks, workspace agents, and interactive apps. For operators, the check is no longer just model quality; it is which stack gives you the cleanest path from prompt to governed work across your real tools.
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.
|
15/36 · Adaptive Collaborator
ChatGPT refreshes plan features with GPT-5.6 models and new workspace… — ChatGPT’s pricing/feature comparison was broadly rewritten to add or update many capabilities across plans, including GPT-5.6 model references, GPT-5.6 Terra/Sol, scheduled tasks, workspace agents, interactive apps, deep research apps, and…
Source: the tool's own site — detected 2026-08-04 by agentic.ai's daily recrawl.
|
|
11/36 · Guided Assistant
Manus now says it is part of Meta and has been repositioned around business… — The site now includes an explicit banner stating that Manus is part of Meta, and the homepage has been rewritten around launching business applications without engineering resources.
Source: the tool's own site — detected 2026-08-03 by agentic.ai's daily recrawl.
|
|
7/36 · Guided Assistant
Sudowrite unveils Muse 1.5, a fiction model that pays royalties to authors — Sudowrite’s homepage now highlights Muse 1.5 as its AI model built specifically for fiction and says it pays royalties to authors.
Source: the tool's own site — detected 2026-08-04 by agentic.ai's daily recrawl.
|
|
18/36 · Adaptive Collaborator
Cursor expands CLI, SDK, and MCP support with new interactive session docs — This looks like a major platform update: Cursor added new CLI installation and interactive-session docs, expanded SDK surfaces for streaming and interaction updates, and broadened MCP/cloud-agent configuration.
Source: the tool's own site — detected 2026-08-04 by agentic.ai's daily recrawl.
|
|
3/36 · Reactive Tool
Quizlet Q-Chat page shifts toward broader study app and teacher tools — The page no longer reads like a Q-Chat-specific product page; it now emphasizes Quizlet’s broader study workflow, practice tests, study guides, and teacher-focused features.
Source: the tool's own site — detected 2026-08-04 by agentic.ai's daily recrawl.
|
|
16/36 · Adaptive Collaborator
Claude adds Cowork, MCP connectors, and Microsoft 365 integrations across… — The Claude site now highlights a broader product lineup and integration surface, including Claude Cowork, Claude Code, Claude Design, Claude Science, Claude for Microsoft 365, Chrome, Slack and Google Workspace connectors, and remote MCP s…
Source: the tool's own site — detected 2026-08-04 by agentic.ai's daily recrawl.
|
What's new in the directory
|
19/36 · Domain Specialist
KaroX turns a local Git repository into a controlled workspace for AI coding clients. It is aimed at developers who want AI to inspect code, run checks, and make approved changes without giving it unrestricted access.
Source: published listing, rubric-scored on ingest.
|
|
13/36 · Guided Assistant
Memora is a state layer for federated learning, multi-agent systems, and edge swarms that aims to keep shared data byte-identical across heterogeneous nodes. It focuses on signed writes, deterministic aggregation, and Byzantine-robust consensus rather than general-purpose agent execution.
Source: published listing, rubric-scored on ingest.
|
|
10/36 · Guided Assistant
Pilot Protocol runs a local daemon and CLI for networked agent coordination, with trust handshakes between nodes and real-time event streaming to your webhook endpoint.
Source: published listing, rubric-scored on ingest.
|
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.
|
MAXMEL Tech Ltd.
product launch
MAXMEL Tech Ltd. says WEXTL has moved from closed beta into full public production. The visual workflow builder is aimed at businesses that want to deploy AI agents at scale without runaway automation costs. For those tracking AI agent platforms, this signals another push to package agent building for nontechnical teams and to compete directly with established workflow tools. The key question is whether WEXTL can really deliver predictable scaling costs where others have struggled.
Via Agentic-News.ai — Barchart.com. Read: https://www.barchart.com/story/news/3588353/wextl-announces-official-public-launch-an-ai-agent-builder-and-visual-workflow-automation-platform
|
|
Stairwell
product launch
Stairwell announced Backstory on July 29, 2026 as an agentic investigation platform for malware response. The launch follows analysis showing 2.4 related variants for every sample named in public threat reports. For those tracking agentic AI, this signals security tools shifting from alerts to autonomous investigation. The key promise is how fast Backstory can identify affected systems and connect the dots across an attack.
Via Agentic-News.ai — The Manila times. Read: https://www.manilatimes.net/2026/07/29/tmt-newswire/globenewswire/stairwell-launches-backstory-the-first-agentic-investigation-platform-for-malware-blast-radius/2394146
|
|
BrowserStack
product launch
BrowserStack launched Test Companion on July 29, 2026 as agentic AI for test automation inside the IDE. The tool is built to speed up test authoring, execution, debugging, and maintenance across web and mobile apps. For those tracking agentic AI, this signals testing becoming a bigger target for autonomous workflows. The detail to watch is whether moving testing into the IDE finally closes the gap between code and release.
Via Agentic-News.ai — FinanzNachrichten.de. Read: https://www.finanznachrichten.de/nachrichten-2026-07/69156897-browserstack-launches-test-companion-agentic-ai-that-brings-complete-test-automation-into-the-ide-008.htm
|
Biggest movers this week
Listings with the largest week-over-week jump in outbound-click interest
- Claude Code · +86.4% 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.
- OpenHands · +36% 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).
Biggest declines this week
Listings with the largest week-over-week drop in outbound-click interest
- Cline · -44% 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.
- 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).
Wingman is a messaging-first autonomous AI agent by Emergent. It connects to your email, calendar, Slack, CRM, and GitHub, then handles scheduling, research, sales support, and admin tasks through WhatsApp and Telegram conversations.
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 bot maker for discord"
- "best coding ai"
- "Agentic AI to analysis files from PDFs files"
- "AI agent to triage and reply to my email inbox"
- "best coding agent for a large existing codebase"
Learn: How Refinery actually works
Refinery is not an agent that plans trades or takes actions on its own. Based on the description, it packages DeFi and network intelligence into JSON bundles that other agents or trading systems can read. In plain English, it is a data layer, not an operator. Its job is to fetch and structure market context so a downstream system can decide what to do next. The 4/36 score fits that picture: little evidence of autonomous behavior, state management, or session-level control. It may help a workflow stay informed, but it does not appear to keep the loop closed. The human or the trading system still makes the calls and executes them.
See Refinery's full scorecard →
One tool worth testing
Today's signal didn't produce a clear single-tool pick worth a test session. Tomorrow's read picks it back up.
Our pick: n8n
The use case: workflow automation.
Tops our independent /36 — strongest on action and reliability, ahead of Zapier. Pricing: Per-execution (not per-step).
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 →
|