Player profile
Raven
An Apache-2.0 host agent that plans complex tasks as DAGs and orchestrates built-in research, coding, design, and on-call agents plus third-party agents such as Claude Code, Codex, OpenClaw, and Hermes Agent over ACP, CLI, or OpenAI-compatible APIs.
Profile
Where it fits
Raven is an Apache-2.0 host agent from EverMind that brings built-in and third-party agents together to carry out complex tasks. Its built-in agents are Raven-Research for autonomous deep research with structured reports and traceable sources, Raven-Code for agentic software development, debugging, refactoring, and data analysis, Raven-Design for slide decks, brand assets, charts, diagrams, and web interfaces, and Raven-Oncall for unattended workflow automation over hours or overnight.
It belongs in Multi-Agent Platforms And Builders because orchestration is the product surface: Raven generates a DAG for a complex task, coordinates specialized agents, and manages task dependencies and parallel execution. Third-party agents connect over ACP, CLI, or OpenAI-compatible APIs, with presets for Claude Code, Codex, OpenCode, Hermes Agent, OpenClaw, MiroThinker, GitHub Copilot, Qwen Code, CodeBuddy, Qoder, Grok Build, Kimi Code, and Pi.
Editors and hosts can launch Raven itself as an ACP server; it also serves A2A tasks over HTTP JSON-RPC and uses MCP for external tools and resources. EverOS memory preserves user context, agent experience, and world knowledge across sessions. Raven ships a WebUI, a terminal UI, and a Docker setup, and is pre-alpha, so interfaces may change.
Builder web analytics
Measure projects built with Raven
Raven's built-in coding and design agents can build software changes and web interfaces as part of a larger task DAG. For a separately instrumented project, Agent Analytics can report how the shipped surface performed so the next Raven task starts from measured user behavior.
This is an optional measurement workflow for projects built with Raven, not a built-in Raven integration or a claim that Raven reports to Agent Analytics. Instrument the deployed surface separately, and configure Agent Analytics access only in the Raven setup that should query it.
First loop to measure
- a user asks Raven to plan and build a website, docs path, onboarding flow, or product change, and Raven runs it as a DAG across built-in or third-party agents
- the reviewed change is deployed to a project that is separately instrumented with Agent Analytics
- the deployed surface reports visits, sources, CTA clicks, signups, activation, and funnel events to Agent Analytics
- a later Raven task with Agent Analytics access configured reads those results and scopes the next change from measured outcomes
Copyable prompt
Use Agent Analytics for this project. If event reporting is missing, add the tracker and report events for this project surface, including Raven-built page, docs path, traffic source, CTA click, signup, activation event, funnel step, or shipped change. Verify events are arriving. Then fetch the last 7 days and compare them with the prior 7 days. Tell me which Raven-built page, docs path, traffic source, CTA click, signup, activation event, funnel step, or shipped change moved users toward value, where users dropped off, which sources mattered, and what my agent workflow should improve next.