# Codecast > Codecast gives your team one place to watch, steer, search, and remember every AI agent session — Claude Code, Codex, Gemini, Cursor — on any machine. ## Product - [Codecast — watch, steer, and search every agent session](https://codecast.sh): Codecast gives your team one place to watch, steer, search, and remember every AI agent session — Claude Code, Codex, Gemini, Cursor — on any machine. - [About — Codecast](https://codecast.sh/about): Why we built Codecast: coding agents forget everything between sessions, and teams lose the reasoning behind their own code. Codecast keeps the record. - [Features — Codecast](https://codecast.sh/features): The cast CLI and everything codecast records: a searchable memory of every agent conversation, cast blame from a line to the conversation that wrote it, a live inbox you can steer from anywhere, and the agents you already run. - [Pricing — Codecast](https://codecast.sh/pricing): Free forever for individuals. Team at $20/seat/month (early access). Enterprise on request. Bring your own agent subscriptions — codecast never resells or marks up model usage. - [Download — Codecast](https://codecast.sh/download): Download the Codecast desktop app for macOS. Watch, steer, and search every agent session. - [Changelog — Codecast](https://codecast.sh/changelog): What's new in Codecast: releases across the CLI, web, desktop, and mobile apps. - [Security — Codecast](https://codecast.sh/security): How Codecast protects your data: code stays on your machine, sessions sync over TLS, sharing is opt-in, and you can self-host the backend. - [Support — Codecast](https://codecast.sh/support): Get help with Codecast: setup, syncing, teams, and troubleshooting. - [Codecast vs the alternatives](https://codecast.sh/compare): Honest side-by-side comparisons of Codecast with other coding agent tools — Delta, Conductor, Vibe Kanban, Happy, Claudia — and when each is the better choice. - [Codecast vs Delta — which coding agent tool fits?](https://codecast.sh/compare/codecast-vs-delta): Delta, from the makers of the Zed editor, and codecast share a premise: the conversation with the agent is the unit of work, and the team should see it live. Delta builds that on its own agent and its own file history. Codecast builds it on the agents you already run and on git. - [Codecast vs Conductor — which coding agent tool fits?](https://codecast.sh/compare/codecast-vs-conductor): Conductor runs a fleet of agents in parallel from one Mac app. Codecast records and steers the sessions your whole team runs, on every machine. - [Codecast vs Vibe Kanban — which coding agent tool fits?](https://codecast.sh/compare/codecast-vs-vibe-kanban): Vibe Kanban plans and dispatches agent tasks from a board. Codecast is the record and memory of every session your team's agents run. - [Codecast vs Claude Code Remote Control — which coding agent tool fits?](https://codecast.sh/compare/codecast-vs-claude-code-remote-control): Anthropic ships remote control for your own live Claude Code sessions. Codecast records every session your team runs, across four agents, and keeps them after they end. - [Codecast vs Happy — which coding agent tool fits?](https://codecast.sh/compare/codecast-vs-happy): Happy is a polished remote control for your own Claude Code sessions. Codecast is a team-wide record, memory, and steering layer for every agent. - [Codecast vs Claudia — which coding agent tool fits?](https://codecast.sh/compare/codecast-vs-claudia): Claudia is a desktop GUI that wraps Claude Code on your machine. Codecast leaves your terminal alone and syncs every session to a team dashboard. ## Documentation - [Documentation — Codecast](https://codecast.sh/documentation): Install the cast CLI, sync every agent session to one searchable place, and give your agents shared memory. Setup, commands, and deep guides. - [How agent snippets work — Codecast docs](https://codecast.sh/documentation/agent-snippets): cast install writes versioned instruction sections into your agents' own config files. This is the mechanism every other guide builds on. - [Give Claude Code memory across sessions and teammates — Codecast docs](https://codecast.sh/documentation/memory): Not notes files: every session can search, read, and watch every other session your team has run. The commands, the scopes, and how agents use them. - [How to find which AI agent session wrote a line of code — Codecast docs](https://codecast.sh/documentation/which-session-wrote-this-line): git blame names whoever committed a line. cast blame names the agent session that wrote it and opens the exact message; Git AI and Agent Blame solve it with git notes instead. - [Messaging between sessions — Codecast docs](https://codecast.sh/documentation/messaging): cast send turns sessions into teammates: any session can message any other, including a teammate's, and manage what the human sees in the inbox. - [Ambient awareness — Codecast docs](https://codecast.sh/documentation/ambient-awareness): Stable mode injects a live feed of recent sessions into every new session at start. Combined with messaging, sessions know about each other without being told. - [See your whole team's Claude Code sessions in one place — Codecast docs](https://codecast.sh/documentation/team-sessions): Claude Code already writes every session to disk. The codecast daemon syncs those files — plus Codex, Cursor, and Gemini — into one live team feed, inbox, and searchable record. - [How to share a Claude Code session with your team — Codecast docs](https://codecast.sh/documentation/share-a-session): Three different asks hide behind that sentence: read a finished conversation, watch a running one, or make every session visible by default. What Anthropic ships, what Lore does, and where codecast fits. - [Pinned thread state — Codecast docs](https://codecast.sh/documentation/thread-state): cast state keeps one agent-written line saying where a thread stands, pinned above the composer and on the inbox card, with its staleness on show. - [Decisions: asking without interrupting — Codecast docs](https://codecast.sh/documentation/decisions): cast decide puts a question, its options and the reasoning into a queue you clear when you choose to. The answer returns to the agent as a message. - [Team chat that agents take part in — Codecast docs](https://codecast.sh/documentation/team-chat): Channels, threads and direct messages where a mention can wake a role or a session, agent lines are capped, and a Slack workspace mirrors in. - [Huddles and walkie — Codecast docs](https://codecast.sh/documentation/calls): Every huddle is transcribed with exact speaker attribution and leaves a digest, so an agent can quote what was said on the call. - [Forks and spawned sessions — Codecast docs](https://codecast.sh/documentation/forks-and-spawn): cast spawn --subagent delegates a worker that nests under the session that launched it; plain cast spawn and cast fork start independent threads in the human's inbox. - [Tasks and plans — Codecast docs](https://codecast.sh/documentation/tasks-and-plans): The work tracking layer agents report into: tasks, plans, binding, comments, and the dashboard that watches it all. - [Triggers — Codecast docs](https://codecast.sh/documentation/triggers): Follow-up work that runs after the session ends: delayed, recurring, or fired by a GitHub event. - [Workflows — Codecast docs](https://codecast.sh/documentation/workflows): Execution graphs in DOT syntax: agent steps, shell commands, conditions, and human approval gates. - [Orchestration — Codecast docs](https://codecast.sh/documentation/orchestration): A conductor agent decomposes a plan, spawns implementers in isolated worktrees, and runs reviewers and critics over the result. - [Pull requests and issues as codecast objects — Codecast docs](https://codecast.sh/documentation/pull-requests): cast pr and issue sync keep a copy of GitHub and Linear objects current from webhooks, send every action back, and wake the session that owns the work. - [The org: roles, scopes and the line — Codecast docs](https://codecast.sh/documentation/org-roles): Route work to a standing responsibility instead of a session: roles with scopes, wakes, proposals a person accepts, and a line with independent review. - [Driving the human's own Chrome — Codecast docs](https://codecast.sh/documentation/browser): cast browser works in a background tab of the Chrome that already holds your logins, and puts the evidence in the thread. - [Driving a native macOS app — Codecast docs](https://codecast.sh/documentation/computer): cast computer reads a window as an indexed tree, refuses stale indexes, and reports whether an action was verified. - [One typecheck watcher for every session — Codecast docs](https://codecast.sh/documentation/typecheck): cast check answers every session from one tsc --watch for each tree and project, so thirty agents do not build the same program thirty times. - [The cast-* skills — Codecast docs](https://codecast.sh/documentation/skills): 23 packaged procedures, compiled into the CLI, each a fixed sequence of ordinary cast commands. - [Sessions on machines you are not sitting at — Codecast docs](https://codecast.sh/documentation/remote-and-cloud-sessions): How a session starts on, moves to, sleeps on and is watched from another machine, and what each lease does when the machine goes away. - [Usage limits are a pause — Codecast docs](https://codecast.sh/documentation/usage-limits): Codecast parks a session that hits a limit, then continues it at the reset or on a saved account that still has room. - [The visual canvas — Codecast docs](https://codecast.sh/documentation/visual-canvas): Agents reply with sandboxed HTML that renders inline: charts, dashboards, diagrams, and small widgets instead of ASCII art. - [Published pages — Codecast docs](https://codecast.sh/documentation/publish): cast publish turns a file into a page at a stable URL, with version history, access gates, and viewer comments that flow back to the session. - [How the client syncs — Codecast docs](https://codecast.sh/documentation/sync-engine): Every surface paints from a local store, an append only log for each scope delivers only what changed, and one window syncs while the others copy it. ## Blog - [Blog — Codecast](https://codecast.sh/blog): Notes on agent memory, attribution, and steering coding agents at team scale. - [The pull request that knows its sessions — Codecast](https://codecast.sh/blog/the-pull-request-that-knows-its-sessions): When agents write most of the code, a pull request is the end of a conversation you were not in. Codecast keeps the two attached: every PR carries its checks, its reviews, and the sessions that made it, and a review can wake the agent that owns it. - [Agents that talk to each other — Codecast](https://codecast.sh/blog/agents-that-talk-to-each-other): A Claude Code agent shipped a change that silently stalled 27 transcripts. A Codex agent in the same checkout found it, fixed it, and told the first one. No human relayed a word. - [A URL for everything your agent makes — Codecast](https://codecast.sh/blog/a-url-for-everything-your-agent-makes): Reports, dashboards, design proposals — agents produce them daily, and chat transcripts bury them. cast publish turns a file into a live page with versions, comments, and a link you can actually send. - [This post wrote itself (on a schedule) — Codecast](https://codecast.sh/blog/this-post-wrote-itself): Codecast triggers run full agent sessions on a timer. The proof is this blog: last week's post and this one were both written, unattended, by runs of the same weekly trigger. - [Your agents forget. Your team doesn't have to. — Codecast](https://codecast.sh/blog/your-agents-forget-your-team-does-not): Every agent session is a problem being solved out loud, and then the terminal closes. Codecast keeps the record searchable — so the next agent, or the next person, starts from the answer. - [An inbox for your agents — Codecast](https://codecast.sh/blog/an-inbox-for-your-agents): Agents don't fail loudly. They finish, or stall, and wait for you to notice. The inbox makes the waiting visible — every session, every machine, sorted by who needs you. - [git blame for AI agents — Codecast](https://codecast.sh/blog/git-blame-for-ai-agents): When an agent writes the line, the author column goes blank. cast blame fills it back in — with the conversation that wrote it. ## For AI agents If you are a coding agent (or assisting a developer) and want to set up Codecast: - Install the CLI non-interactively: `curl -fsSL https://codecast.sh/install | sh` (macOS/Linux) or `npm install -g @codecast-sh/cli`. Homebrew: `brew install codecast-sh/tap/codecast`. - Then run `cast login` — it prints a browser URL for the human to approve; no credentials pass through the agent. - After setup, a background daemon records the agent sessions already running on the machine (Claude Code, Codex, Cursor, Gemini). Nothing about how sessions are run needs to change. - Agents gain team memory: `cast search ""` and `cast ask ""` query every past session, and `cast blame :` traces code to the conversation that wrote it. - Codecast is free for individuals and MIT-licensed (https://github.com/codecast-sh/codecast); the backend is self-hostable.