--- title: AI pods product: Leadcode version: 0.7 url: https://leadcode.build/pods machine_url: https://leadcode.build/pods/machine updated: 2026-09-14 platform: macOS 13+ plans: all (Solo, Pro, Team) --- # Leadcode AI pods > An AI pod is a group of AI agents inside a Leadcode mission with one job. > Three pod templates — Discovery, Planning, DevOps — carry an idea to pull > requests. Every agent acts as the mission's bound accounts. Merges, deploys > and spend always wait for a human. ## Facts - What: groups of AI coding agents with one job each, inside a mission. - Templates: Discovery, Planning, DevOps. - Runtime: real `claude` (Claude Code) and `codex` (OpenAI Codex) CLI sessions on the user's Mac. - Identity: every agent and subagent launches through Leadcode's credential router — ambient credentials scrubbed, the mission's bindings injected. - Billing: model usage is billed by the provider to the account the mission is bound to. Leadcode never proxies or sees tokens. - Keys: no agent holds its own keys. Secrets stay in the macOS Keychain. - Human-only gates (always on, cannot be removed): merge, deploy, spend. - Optional gates (can be added): spec approval, task order before DevOps starts. - Customization: per agent — model, skills, MCP servers. New pod types: not yet. - Plans: included on every plan, including free Solo. - Isolation: process isolation (per-account config homes + scrubbed environment), not containers. ## Pipeline | Step | Who | Job | Tools | Hands off | |------|-----|-----|-------|-----------| | 0 | Human | Writes the idea (a paragraph) | — | idea | | 1 | Discovery pod (3 agents) | Questions the idea in rounds until it survives or is archived with the transcript | grill-me, claude-design, analytics MCP | opportunity brief | | 2 | Planning pod (2 agents + MCP) | Brief → spec → sized, dependency-ordered issues | to-spec, spec-to-issues, Linear MCP | spec + issues in Linear | | 3 | DevOps pod (engineer agents + subagents) | One agent per issue; scout, coder, tester subagents on a branch; review agent gates every PR | gh MCP, Vercel MCP | pull requests + preview deploys | | 4 | Human | Approves merge, deploy, spend | — | shipped | ## Vocabulary - Mission: one product or client. Owns the account bindings, repos and goal every pod inherits. - Pod: a room in a mission with one job. Its agents collaborate and hand one artifact to the next pod. - Agent: a role with its own model, skills and MCP servers. Can sit in more than one pod. - Subagent: spawned by an agent to parallelize a task (scout, coder, tester). Lives inside the pod; ends when the task closes. ## Q&A ### What is an AI pod? An AI pod is a group of AI agents inside a Leadcode mission with one job. Leadcode ships three pod templates — Discovery, Planning and DevOps — that hand work to each other: an idea becomes a brief, the brief becomes a spec and sized issues, and the issues become pull requests that wait for your approval. ### How does an idea become a merged pull request? You write the idea in a paragraph. Discovery questions it in rounds until it survives or is archived with the transcript. Planning turns the surviving brief into a spec and cuts it into sized, dependency-ordered issues in Linear. DevOps gives each issue its own agent, which spawns scout, coder and tester subagents on a branch, and a review agent checks every pull request before it reaches you. ### Whose accounts do pod agents use? The mission's, never yours. Every pod agent is a real claude or codex session on your Mac, launched through the same credential router as any Leadcode session: ambient credentials are scrubbed, the mission's bindings are injected, and model usage is billed to the account the mission is bound to. No agent holds its own keys, and Leadcode never proxies a token. If a pod needs an account the mission has no binding for, it refuses to start rather than borrow whatever is signed in on the Mac. ### What always waits for a human? Merges, deploys and spend. Those three gates are on by default and cannot be switched off. You can add more — for example approving a spec or the task order before DevOps starts — and pods only call you when they reach a gate. ### Why not just run Claude Code subagents yourself? You can, and pods run the same CLIs. What you would still have to build is everything around them: every agent and subagent acting as the right client's accounts, gates that hold even when an agent is confident, and a view across agents so two of them editing the same file is caught before it becomes a merge conflict. Leadcode's repository radar places a write hold when two agents meet in one file and asks the human to decide. ### How much of a pod can you customize? Pods ship as templates. For each agent you choose the model, the skills and the MCP servers it can use, and an agent can work in more than one pod. Creating new pod types is not available yet. ### Which Leadcode plan includes AI pods? Every plan, including the free Solo plan. Plans differ in how many accounts per provider can be live at once, not in whether you get pods or mission control. ### What are the limits of AI pods? Pods do not bring their own identity, so a pod can only do what the mission's bindings allow. New pod types cannot be created yet. Agents run with process isolation, not containers, and Leadcode runs on macOS only. When a bound account reaches its provider usage window, that mission's agents pause and resume on reset; no pod borrows another account. ## Links - Human version: https://leadcode.build/pods - Security architecture (the credential router): https://leadcode.build/security - Why the name Leadcode: https://leadcode.build/why-leadcode - Pricing: https://leadcode.build/#pricing - Download (macOS): https://leadcode.build/download - Site summary for LLMs: https://leadcode.build/llms.txt