mpai lets a named teammate view an explicitly shared native Codex or Claude Code session and contribute attributed prompts from another terminal. The host keeps the running session, with private Tailscale networking, roles and revocation.
claudemon is a local terminal game connected to Claude Code activity hooks. Prompt length and working time create encounters in a second terminal, while the companion shows agent activity and completion or approval notifications.
AgentSky runs a marketplace of coding and task agents in the browser or through an API. Users choose a harness and model, compare agents and optionally connect eligible existing provider subscriptions or accounts.
Vibe Buddy is a prototype desk display for Codex and Claude Code usage limits, active task counts and attention states. A local companion reads tool activity and sends display information to the device over Bluetooth.
hotcell provisions agent sandboxes on owned hardware through containers or supported microVM runtimes. A host gateway can replace provider credentials with scoped sandbox tokens, while fleet monitoring, admission control and optional egress policies manage execution.
Atlaso captures decisions, preferences and lessons into personal and project memory that connected assistants can recall. Users inspect, correct or forget memories, with Pro adding cross-tool/device sharing, refinement and source-backed memory questions.
Exabase extracts public web pages, including JavaScript-rendered content, into Markdown or JSON through an asynchronous API. The same extraction endpoint supports documents and media, with polling or Markdown webhooks for completed jobs.
CopilotKit connects agents to product interfaces through AG-UI, chat and generative UI components, frontend actions and shared state. Developers can use ready-made components or headless APIs, with hosting and enterprise deployment options.
Brandfetch provides searchable brand logos, colors, fonts and company context through web search and APIs. Brands can claim their profiles, while developers use separate logo, search and full-brand data endpoints to enrich applications and agent workflows.
Troopr drafts engineering updates from Jira, GitHub and Slack activity, gathers async or live standup context and posts reports. It supports project issue actions, recurring routines, retrospectives and proposed Jira updates for confirmation.
Superbrain indexes repositories with TokenFold and retrieves selected context for coding tasks. It plans edits, runs checks and prepares pull requests, with proposed edits and commands presented for developer approval.
Progress Agent Engineering traces model calls, prompts, retrieval and tools, then connects costs and failures to evaluations and experiments. Native .NET, Python and TypeScript support helps teams investigate production agent behavior and review governance controls.
Merge assesses candidates through a pull-request review loop in a realistic small codebase. Candidates identify bugs, refactors and risks; an AI agent responds with revisions, while reports summarize judgment, priorities and estimated token/cost efficiency.
HAR coordinates coding workflows through a repository-owned harness contract, isolated worktrees and launch/verify/teardown stages. Its CLI and MCP expose project checks and evidence artifacts, while Mission Control tracks runs, validations and repositories.
Crew displays small animated creatures for coding-agent chats and subagents, reflecting lifecycle states and blocked sessions. Local hook/status files drive the display; Pro on Mac adds session jumps, approval controls and token/cost dashboards.
Annotate records screen walkthroughs with drawings, gestures and spoken instructions synchronized to frames. Sessions can be handed to existing coding agents, which read the recording context through MCP rather than only a static screenshot.
Remix gives teammates live sandbox variants of an existing codebase to edit with their AI coding tools. It records prompts, agent responses and file changes, then sends engineering a reviewable request with a preview link and configured guardrail results.
AgentConnect brings multiple coding or ACP agents into team channels with roles, handoffs, schedules, tool access and shared visibility. A console manages sessions and permissions, while daemons run agents in selected execution environments.
Trigger.dev runs TypeScript tasks and durable AI agents with retries, queues, streaming and observability. Its chat-agent workflow supports paused tool approvals and continued execution, while deployments version task code and Realtime exposes run status.
Statewave records immutable episodes, compiles typed memories and retrieves ranked context within a token budget. It adds source provenance, sensitivity labels, configured policies and integrity receipts, backed by self-hosted Postgres and optional connectors.
Paritok sits between coding agents and model providers, filtering tool schemas, compressing file/tool output and summarizing older history. Modified segments retain references so agents can retrieve original content when needed.
Bullet is a coding agent with model routing, targeted repository search and parallel independent tool calls. Its desktop app and CLI use the same agent loop, with interception of duplicate calls and stuck execution loops.
bb brings installed coding agents into a local workbench with separate threads, worktrees and plugins. Agents can customize panels and commands, while its CLI allows scripts or scheduled programs to start tasks in the workspace.
agent-manager is a Go terminal interface over tmux that groups installed coding-agent sessions by project. Users can spawn or answer agents, inspect repository diffs, send line comments and resume supported conversations from one session list.
Free
API available
Checked Oct 9, 2026
Showing 168 of 2,280 tools
Plan facts carry individual sources and check dates. Unknown values are excluded from required filters.