Terminal native
Tools and execution
Agent runs, Skills, MCP, Hooks, approvals, the sandbox and worktrees stay with Codex CLI or Claude Code.
Loading page
Please wait while the page loads.
Please wait while the page loads.
Terminal protocol bridge
Codex CLI and Claude Code keep their tools, permissions and local execution; DeepCodex adds project context, routing and metering only where the model request passes through.
Pick a terminal to follow its protocol flow
Compatibility contract
DeepCodex does not clone either terminal or take over local tools. It fits the orchestration you already use and enhances the part the model request passes through.
Agent runs, Skills, MCP, Hooks, approvals, the sandbox and worktrees stay with Codex CLI or Claude Code.
Project profile, task memory and hard-task enhancement injected into the request, routed between the fast and deep model by stage.
Every parent task and child request traces back to model, equivalent tokens, cost, TTFT, duration and failure reason.
Feature map
Names follow each terminal's own concepts; DeepCodex describes only what it adds on the request side.
A per-project profile with the stable stack, conventions and pitfalls
Model mix, quota and per-call tracing for concurrent requests
Keeps the terminal flow; picks the fast or deep model by stage
Tools stay connected by the terminal; model requests land in one ledger
Never crosses terminal permissions; adds key, project and budget bounds
Groups concurrency, models, equivalent tokens and results by parent task
Experimental capabilities such as Agent Teams and dynamic workflows depend on the terminal's current version and configuration; DeepCodex does not change the terminal's permission decisions.
Terminal coverage
Each terminal is labelled by its real protocol and configuration; an integration limited by the terminal is never dressed up as full support.
Goal-driven execution
A goal is not a longer prompt; it is a checkable delivery contract with target, boundary, verification and budget settled first.
Orchestration patterns
Different task sizes suit different structures. DeepCodex covers all of them with the same project context and metering.
Hand exploration, implementation, tests or review to an isolated context; the main thread only collects results.
Several tasks advance in separate workspaces so the same files are never edited at once.
Let a script own loops, branches and intermediate results; the model handles only the current step.
Independent agent runs produce separately; a summariser compares the evidence and resolves conflicts.
Keep the agent runs, Skills and tooling you know in Codex CLI or Claude Code, and gain the model mix, project memory and transparent metering.