Follow new capabilities, tutorials, field reports and reviews. Learn what assistants can do, how to use them and where their limits lie.
MCP accessRead-onlyConnect your agent to BotClaw
Connection guideSkills expose a directory first and load detailed instructions or references only when needed. The documentation also covers installation and /learn. Skill files remain editable; review their source and execution permissions before using them.
Website review date, not publication date
One gateway can host multiple agents with separate workspaces, state and session history. Bindings route incoming messages. The documentation warns against sharing agentDir, making it useful when separating work and personal assistants.
Website review date, not publication date
The official quickstart covers installation, model providers and troubleshooting. Verify an ordinary conversation first, then connect a messaging gateway and add skills, automation or routing incrementally so configuration problems stay easier to diagnose.
Website review date, not publication date
The official automation index explains persistent jobs, schedules, triggers and delivery to chat channels or webhooks. Start with a reminder or recurring brief, then inspect run history rather than assuming a configured job has succeeded.
Website review date, not publication date
The official guide walks through a dedicated WhatsApp number, gateway, workspace and sender allowlist. Establish one working conversation before adding proactive tasks, and review tool permissions and isolation.
Website review date, not publication date
Persistent files and the snapshot injected at session start are different. New entries generally become visible in the next session. The guide recommends fresh sessions at natural task boundaries and checking that a write occurred instead of trusting an acknowledgement.
Website review date, not publication date
The security guide explains approvals, deny rules and timeouts. Shell-command matching is not a complete OS sandbox: containment also requires suitable permissions, restricted mounts and scoped credentials. A single deny rule cannot guarantee that a capability is unreachable.
Website review date, not publication date
The v2026.9.7 release links to reader-friendly notes and a Markdown changelog, with npm package and integrity information. Review changes relevant to your channels and configuration before upgrading. BotClaw has not independently tested this release.
Instinct describes an assistant connected to email, messaging and devices, reached by text or phone and trained to use its own phone and computer. The homepage mentions follow-ups and travel arrangements. It has no publication date; this profile is filed by the date checked, not a launch date.
Website review date, not publication date
Meta's official feed introduces Muse for Small Business as a personal AI agent that helps advance business goals in the background. This summary is based on the RSS introduction. See the source for availability, capabilities and conditions.
Version 2026.8.33 is a gateway-only extended-stable release. It backports model support, security hardening and reliability fixes to the August branch, including metrics authorization and Discord media permissions. Check which release branch your installation uses before upgrading.
OpenAI describes always-on assistants with their own cloud computers, connected apps and memory of user preferences. Read-only proactive research is distinguished from action-taking work, with permissions and approval rules. Availability is rolling out in eligible markets; check the official page for current plan requirements.
Cue is a standalone personal-agent app from Manus. Its announcement describes agents with their own email, phone number, wallet and computer, working within a user-set budget and collaborating on shared goals. At launch it is invite-only early access, with the iOS app awaiting store review.
Team Bots combine shared context, plugins, credentials and memory, including collaboration in Slack. The announcement says each member's conversations remain private, with separate personal context and memories. Check the official documentation for permissions and availability.
This stable tag collects changes since 0.21.4 for deployments including Docker and Hermes Cloud. The notes mention the desktop plugin SDK, connectors and performance improvements, while a fuller feature write-up is deferred to 0.22.0. Review the changelog for your deployment.
The official recruiting team describes interview preparation, scorecards, sourcing and process analysis across its ATS. This is internal experience, not evidence that candidate assessments become more accurate or unbiased. Document organisation is a narrower starting point.
An official post-sales author combines a coordinating bot, account specialists and routines to gather context across calls, Slack, email and calendars. The focus is meeting preparation and keeping promises visible, while people retain ownership of customer relationships.
An official author describes an inbox manager, calendar assistant, intelligence scout and coordinating bot. The guide offers a pattern for turning scattered requests into ongoing work. Its productivity impressions belong to the author's own environment.
An official SDR author uses Cursor for account research and first-contact drafts, then Grok Bot to coordinate sales tools and stage follow-ups. CLEAR / FLAG intake and human sending distinguish automatic preparation from external execution.
The official legal-operations guide covers counsel agendas, Slack triage, contract workflows and executive-update drafts. Every outbound message receives human review. It offers document and task-management patterns, not legal advice.