AgentHippo
The AgentHippo platform
Build agents. Ship them anywhere. Then govern them in production.
Every engine in one IDE and CLI — Claude Code, Codex, LangChain, or your own. Any model: cloud, local, or self-hosted. Pack it, ship it anywhere, trace every run.
› See it in the product
One surface for creating, evaluating, scheduling, and managing agents.
› The pieces
Cursor and Claude Code are single-engine seats. AgentHippo is the system that builds agents on Claude Code, Codex, LangChain, or your own engine — and the control plane that runs them in production: routing, budgets, leases, and evidence across channels and providers.
Workbench
Work through agents, not just build them. Claude Code, Codex, LangChain, or your own engine — all in one IDE with instant diffs, file trees, and full traces. No more alt-tabbing between terminals and editors.
Multi-engine IDE →Spotlight
Ask agents to analyze your data or their own traces — and get interactive charts, reports, and dashboards back all in one tool. Debug cost spikes. Spot regressions. Agents do the analysis, you review results.
Agent-powered analysis →Fleet
Deploy, schedule, and govern agents across channels and teams — leases, kill switch, spend caps, and a verifiable action record. IDE for development. CLI for APIs. Gateway for Slack, WhatsApp, and more.
Agent Anywhere →Agent Packs
A 3-layer agent schema: Pack, Engine, Model. Compare how a pack behaves on Codex vs Claude Code, or across models. Versioned, signed, reproducible — the unit of deployment and of rollback.
Why packs →Learning Loop
Compare prompt versions, pack versions, and models. Run evals, detect regressions, promote better variants. Agents improve with evidence, not guesswork.
Scheduled & improving agents →Agent Store
Curated agent packs your org can adopt without rebuilding the same glue — plus your private registry for internal packs, prompts, and skills.
Browse the store →› agenthippo CLI — packages, not scripts
Serve agents in headless mode. Automate agent workflows in terminal or CI. With agenthippo, you run the same versioned Agent Packs by name (for example --agent support-triage) that your team uses in the editor, so local runs, pipelines, and lightweight services stay consistent.
# Interactive session (same engines & models as the IDE)
agenthippo chat --workspace .
# One-shot task
agenthippo ask "Summarize this repository" --workspace .
# Serve an agent (versioned artifact, not a loose script)
agenthippo serve --agent support-triage --workspace . --port 3000
Optional gateway bridge: expose the same pack over Telegram, WhatsApp, and more without a separate integration script. For every subcommand and flag, use agenthippo --help.
agenthippo serve --agent support-triage --workspace . --gateway-bridge --bind telegram,whatsapp
# Discover everything else
agenthippo --help› Schedule agents. Observe the fleet.
Schedule from the UI or CLI, then manage agents in the fleet view. Ask AgentHippo to build your own Spotlight dashboard on top.
› Spotlight: analyze and visualize with agents
Ask an agent data analyst to plot charts, build dashboards, and explain cost — all on your computer — then share the results with your team in one click.
› Agent packs: share work in the agentic era
Fully reproducible, with one-click publish and deploy of the entire agent. Author packs and skills, exercise them on any agent engine in the UI or CLI, then hand off — teammates install in one click from your library or the store.
› Any agent, any model
Shorter eval cycles: compare engines and models in one surface, UI and CLI, instead of standing up parallel toolchains.
› Drop-in. No vendor lock-in.
› Built for every seat at the table
Agent developers
One IDE for every engine, instant diffs, full traces of every run — and any model, cloud or local. Stop juggling terminals; start comparing engines side by side.
Engineering leads
One registry and pack format for the whole team, cost truth per member and per pack, and versioned rollback when a change regresses. No more five frameworks and zero visibility.
Platform & DevOps
A CLI built for CI, scripted deploys that verify themselves, and a fleet view of every scheduled agent — what ran, what failed, what it cost.
Data & analytics teams
Spotlight turns your own trace data into charts and dashboards on request — DuckDB under the hood, everything local-first, shareable in one click.
› Sound familiar?
Every card is a moment we built the platform for.
"Finally I can see where my agent spend goes. Spotlight panel + traces in one place."
"Using Claude and Codex in the same flow with one observability layer. Game changer."
"Local-first traces and cost. No cloud sign-up. Exactly what we needed for our team."
"Finally I can see where my agent spend goes. Spotlight panel + traces in one place."
"Using Claude and Codex in the same flow with one observability layer. Game changer."
"Local-first traces and cost. No cloud sign-up. Exactly what we needed for our team."
"The agent-built charts in Spotlight are exactly what I asked for. No fixed dashboards."
"Cut our API spend by seeing tool loops and prompt bloat. AgentHippo made it obvious."
"The agent-built charts in Spotlight are exactly what I asked for. No fixed dashboards."
"Cut our API spend by seeing tool loops and prompt bloat. AgentHippo made it obvious."
Build free today. Govern when you ship.
Bring your own keys, run any engine, keep every trace local. When your agent earns production authority, the agent control plane is one deploy away.