Taxonomy — 28 capabilities · 112 systems # Agent capabilities Capabilities are atomic tags that describe what an agent can actually do — run code, edit files, drive a browser, take phone calls — independent of which product category it belongs to. A CLI agent and a cloud agent can share 'long-horizon tasks'; a voice agent and a digital employee can share 'human handoff'. The tags let you compare across categories rather than only inside them. Tags verified against the catalog seed of 2026-09-25. Filter the [agent index](https://www.agentlist.io/systems) by capability to compare across categories. F.01 ## Agent capabilities — What the agent itself does [## Background runs /background-tasks Asynchronous cloud or background execution you can walk away from. 10 agents →](https://www.agentlist.io/capabilities/background-tasks) [## Browser operation /browser-ops Drives a real web browser: navigate, click, fill, extract. 12 agents →](https://www.agentlist.io/capabilities/browser-ops) [## Cited answers /citations Returns answers grounded in identifiable sources. 08 agents →](https://www.agentlist.io/capabilities/citations) [## Code execution /code-exec Runs code in a real environment or sandbox. 16 agents →](https://www.agentlist.io/capabilities/code-exec) [## Code review /code-review Reviews diffs and pull requests with feedback. 04 agents →](https://www.agentlist.io/capabilities/code-review) [## Desktop control /desktop-use Pixel-level computer use — screenshots, mouse, keyboard. 03 agents →](https://www.agentlist.io/capabilities/desktop-use) [## File editing /file-edits Reads and writes project files, including multi-file edits. 35 agents →](https://www.agentlist.io/capabilities/file-edits) [## Git & PRs /git-pr Branches, commits, and pull-request workflows. 09 agents →](https://www.agentlist.io/capabilities/git-pr) [## Human handoff /human-handoff Escalates, transfers, or requests approval from a human. 28 agents →](https://www.agentlist.io/capabilities/human-handoff) [## Incident response /incident-response Alert, on-call, and incident-triage loops. 07 agents →](https://www.agentlist.io/capabilities/incident-response) [## Long-horizon tasks /long-horizon Unattended multi-step work running minutes to hours. 26 agents →](https://www.agentlist.io/capabilities/long-horizon) [## Model routing /model-routing Selects or routes across multiple underlying LLMs. 13 agents →](https://www.agentlist.io/capabilities/model-routing) [## Persistent memory /memory Remembers context across sessions or runs. 12 agents →](https://www.agentlist.io/capabilities/memory) [## Repo context /repo-context Indexes or reasons over a whole repository, not just open files. 34 agents →](https://www.agentlist.io/capabilities/repo-context) [## Shell access /shell Executes terminal/shell commands. 25 agents →](https://www.agentlist.io/capabilities/shell) [## Voice I/O /voice Realtime voice or telephony input/output. 09 agents →](https://www.agentlist.io/capabilities/voice) [## Web research /web-research Searches, browses, and synthesizes information from the web. 17 agents →](https://www.agentlist.io/capabilities/web-research) [## Workflow builder /workflow-builder Visual or no-code composition of agent workflows. 06 agents →](https://www.agentlist.io/capabilities/workflow-builder) F.02 ## Infrastructure capabilities — How it composes into a stack [## Agent identity & auth /identity Identity, credentials, or authorization for agents. 07 agents →](https://www.agentlist.io/capabilities/identity) [## Agent payments /payments Payment rails, wallets, or charging for agent actors. 05 agents →](https://www.agentlist.io/capabilities/payments) [## Agent-to-agent protocol /a2a Speaks or implements an agent interoperability protocol. 08 agents →](https://www.agentlist.io/capabilities/a2a) [## API-native /api-native Designed to be embedded and driven via API or SDK. 70 agents →](https://www.agentlist.io/capabilities/api-native) [## MCP tools /mcp Model Context Protocol client or server support. 24 agents →](https://www.agentlist.io/capabilities/mcp) [## Messaging surface /messaging-infra Email, SMS, or chat channels agents can send and receive on. 10 agents →](https://www.agentlist.io/capabilities/messaging-infra) [## Multi-agent orchestration /multi-agent Spawns, coordinates, or composes multiple agents. 12 agents →](https://www.agentlist.io/capabilities/multi-agent) [## Sandbox runtime /sandbox-runtime Provides isolated execution environments as a service. 02 agents →](https://www.agentlist.io/capabilities/sandbox-runtime) [## Self-hostable /self-hostable Can be deployed on your own infrastructure. 16 agents →](https://www.agentlist.io/capabilities/self-hostable) [## Tool calling /tool-calling Function/tool use as a first-class primitive. 12 agents →](https://www.agentlist.io/capabilities/tool-calling) FAQ ## Frequently asked ### What are AI agent capabilities? Capabilities are atomic tags that describe what an agent can actually do — run code, edit files, drive a browser, take phone calls — independent of which product category it belongs to. A CLI agent and a cloud agent can share 'long-horizon tasks'; a voice agent and a digital employee can share 'human handoff'. The tags let you compare across categories rather than only inside them. ### How is this different from the category taxonomy? Categories answer 'what kind of product is it' — an AI IDE, a cloud agent, a voice agent. Capabilities answer 'what can it do' — execute code, call MCP tools, route between models. Categories partition the catalog; capabilities cut across it. ### What do the two facets mean? Agent capabilities are things the agent itself does — file editing, web research, voice I/O. Infrastructure capabilities describe how the system composes into a stack — API-native, self-hostable, agent payments, sandboxed runtimes. Most infrastructure entries are building blocks rather than end-user agents. ### How are capability tags assigned? Each system is tagged only with capabilities verifiable from its documented feature set in the catalog seed. Tags are not self-reported by vendors and are re-checked when the catalog is regenerated. --- Source: [AI agent capabilities — the atomic taxonomy | agentlist.io](https://www.agentlist.io/capabilities). This is the public page rendered as Markdown; interactive controls require the website.