Start with the job, then see which agents document it. Each task below comes from the vendor's own pages, linked on the agent's record. We list what is documented, not what works well. We have not tested these agents.
You want code written, fixed or migrated, and you will review the result.
What to check first: Existing repository versus building from scratch, where code runs, and who reviews changes.
product
Best fit (documented): Developers delegating repository changes from a terminal.
Documented tasks: Terminal coding workflow; Code migrations and bug fixes.
Poor fit: A no-setup personal assistant for errands.
Vendor source · checked 2026-10-01
product
Best fit (documented): Engineering teams delegating well-scoped, verifiable tickets.
Documented tasks: Writes, runs and tests code; Embedded IDE, shell and browser.
Poor fit: Hard-to-verify projects with no completion criteria.
Vendor source · checked 2026-10-01
product
Best fit (documented): Building apps and related artifacts from natural-language requests.
Documented tasks: Writes code, sets up infrastructure and tests; Plan mode before changes; Checkpoints and rollback.
Poor fit: Assuming a prototype is production-ready without review.
Vendor source · checked 2026-10-01
You want an assistant in your messages that handles inbox, scheduling and reminders.
What to check first: Which accounts it can write to, whether actions need your approval, and what is retained after you disconnect.
product
Best fit (documented): A messaging-first assistant connected to personal apps and devices.
Documented tasks: Text or call interface; Connections to applications and devices.
Poor fit: Connecting sensitive services without checking controls and terms.
Vendor source · checked 2026-10-01
product
Best fit (documented): Messaging-based email, calendar and reminder work.
Documented tasks: Email reading, search and drafting; Calendar scheduling and availability; Reminders and web search; Apple Messages, Telegram, WhatsApp and RCS.
Poor fit: Granting broad access before checking approval controls.
Vendor source · checked 2026-10-01
product
Best fit (documented): Personal administration through iMessage.
Documented tasks: iMessage interface; Email, calendar, documents and reminders; Requested briefings, nudges and follow-ups.
Poor fit: Assuming a marketing summary proves every action asks for approval.
Vendor source · checked 2026-10-01
product
Best fit (documented): Personal reminders, ongoing tasks and artifact work.
Documented tasks: Reminders and ongoing conversation context; PDFs, spreadsheets and other artifacts; App and web access.
Poor fit: Unmonitored actions where errors would be costly.
Vendor source · checked 2026-10-01
product
Best fit (documented): Personal connected-app work with shared ChatGPT context.
Documented tasks: Permitted plugins and shared ChatGPT memory; Private proactive research; Device access subject to device permission.
Poor fit: A user needing per-memory inspection and deletion.
Vendor source · checked 2026-10-01
product
Best fit (documented): People who want several agents with separate identities that hand work to each other, and who will review the final result.
Documented tasks: Personal agents with their own email, phone number, wallet and computer; Group chats where several agents hand work to each other; Single personal assistant for errands, reminders and research.
Poor fit: Anyone who needs published spending limits, retention terms or a public price before connecting accounts; none was established in the source we read.
Vendor source · checked 2026-10-01
You want bookings, calls, purchases or tasks that carry on without you watching.
What to check first: Transaction costs, human escalation, spending controls and what the agent may do unsupervised.
product
Best fit (documented): Real-world coordination and life administration with vendor-assisted escalation.
Documented tasks: Bookings and real-world task coordination; Group chats, calls and emails; Human escalation when AI alone cannot finish; Single-use purchase-card mechanism.
Poor fit: A user requiring verified zero-cost transactions or fully AI-only execution.
Vendor source · checked 2026-10-01
product
Best fit (documented): Personal or team app tasks that continue in a cloud environment.
Documented tasks: Tasks that continue on a cloud computer; Connected app and file work; Routines and proactive messages; Signal, Telegram and WeChat access.
Poor fit: A user who needs every connected-app action to wait for approval without configuring boundaries.
Vendor source · checked 2026-10-01
product
Best fit (documented): Delivering task outputs rather than only answering questions.
Documented tasks: Browser and file-system tools; Presentations and websites; Manus 2.0: Cascade engine, Manus Studio desktop workspace, Cloud Computer, Automations, Remote Control, Video Editor, Game Dev.
Poor fit: Work requiring demonstrated reliability on sensitive actions.
Vendor source · checked 2026-10-01
product
Best fit (documented): People who want several agents with separate identities that hand work to each other, and who will review the final result.
Documented tasks: Personal agents with their own email, phone number, wallet and computer; Group chats where several agents hand work to each other; Single personal assistant for errands, reminders and research.
Poor fit: Anyone who needs published spending limits, retention terms or a public price before connecting accounts; none was established in the source we read.
Vendor source · checked 2026-10-01
You run a support or sales operation and want an agent talking to customers.
What to check first: Handoff to humans, how resolutions are billed, and how you check answers against your own content.
product
Best fit (documented): Customer-facing service and sales workflows.
Documented tasks: Service and sales workflows; Customer knowledge and context.
Poor fit: Choosing a vendor solely from its performance superlatives.
Vendor source · checked 2026-10-01
platform
Best fit (documented): Enterprise agents connected to Salesforce workflows.
Documented tasks: Service and employee support workflows; Agent builder and observability.
Poor fit: Treating enterprise deployment as a consumer assistant.
Vendor source · checked 2026-10-01
You want agents that run on schedules or triggers against company data and tools.
What to check first: Admin controls, data permissions, write-action approvals and workload pricing.
platform
Best fit (documented): Repeatable team workflows with configured access and triggers.
Documented tasks: Schedules and API triggers; Apps and Slack channels; Write-action approvals and parameter constraints.
Poor fit: Assuming every ChatGPT subscription includes workspace agents.
Vendor source · checked 2026-10-01
platform
Best fit (documented): Teams building business-data agents for several channels.
Documented tasks: Business-data connections; Natural-language agent building; Multi-channel publishing.
Poor fit: Comparing it as a ready-made personal assistant.
Vendor source · checked 2026-10-01
platform
Best fit (documented): Enterprise agents connected to Salesforce workflows.
Documented tasks: Service and employee support workflows; Agent builder and observability.
Poor fit: Treating enterprise deployment as a consumer assistant.
Vendor source · checked 2026-10-01
You are a developer who wants to build agent behaviour rather than buy a product.
What to check first: How much you must build and run yourself, persistence, and human-in-the-loop controls.
framework
Best fit (documented): Developers implementing stateful agent workflows.
Documented tasks: Durable execution and persistence; Human-in-the-loop state inspection; Short- and long-term memory support.
Poor fit: A finished assistant that works without development.
Vendor source · checked 2026-10-01
platform
Best fit (documented): Teams building business-data agents for several channels.
Documented tasks: Business-data connections; Natural-language agent building; Multi-channel publishing.
Poor fit: Comparing it as a ready-made personal assistant.
Vendor source · checked 2026-10-01
Pick the group that matches your job, open each agent's record for permissions, pricing status and limits, then compare two with the comparison tool. Pricing and plan eligibility vary, and unknown is not free. See the methodology and the privacy and approval checklist.