Coders Talk for Claude Code and Codex

Send the Claude Code or Codex session you are in to Coders Talk as a draft Build: the prompts, interventions and fails that mattered, with the raw log one layer down. The plugin never publishes. The draft is private: only you see it, or your team when the session ran in one of the team's repositories. You review every moment on the site and publish it there if you want to.

The other way round, the plugin gives your agent the Coders Talk library: before a non-trivial task, or after a few failed attempts, it can look up sessions where other developers did something similar, and what went wrong for them (see The library in your agent). And /coders-talk:use puts a published Build's playbook, what that session learned the hard way, into your repository as a skill or a rule (see Playbooks).

Install

From a terminal

curl -fsSL https://coders.talk/install.sh | sh        # macOS and Linux
irm https://coders.talk/install.ps1 | iex              # Windows, in PowerShell
coders-talk login

This installs coders-talk, one file with no Node.js needed, into ~/.coders-talk/bin and adds that folder to your PATH (the rc file of your shell; on Windows the user PATH). The installer takes the file for your platform from this repository's releases and checks it against the release's SHA256SUMS. No sudo or administrator rights. CODERS_TALK_VERSION=0.11.0 installs that version, CODERS_TALK_NO_MODIFY_PATH=1 leaves PATH alone.

coders-talk login waits in the terminal until you press Connect in the browser; over SSH it prints the link and the code to open elsewhere. coders-talk update replaces the file with the latest release after checking its SHA256. A command you type mentions a newer release in one line, at most once a day; it never waits for the network to do so, and hooks never check. CODERS_TALK_NO_UPDATE_CHECK=1 turns that off.

Then coders-talk enable connects Claude Code and Codex to it. It says what it found and what it will do, asks, and then installs the plugin through each agent's own plugin system: it lays the plugin out in ~/.coders-talk/plugin and adds that folder as the local marketplace coders-talk-local (claude plugin marketplace add, claude plugin install coders-talk@coders-talk-local; codex plugin marketplace add, codex plugin add). It is the same plugin as the one below, but its hooks and skills call coders-talk by its absolute path, so no Node.js is needed and desktop apps find it without your shell's PATH. On the way it:

• replaces the plugin from the GitHub marketplace (coders-talk@coders-talk), whose hooks would otherwise run twice;
• finds a Coders Talk MCP server you added by hand (claude mcp add, mcp_servers in Codex's config.toml) and removes it, or keeps it and installs the plugin without its own server for that agent;
• asks once about auto mode for every agent it found (off unless you say otherwise). Codex runs the hooks only after you trust them in /hooks; enable does not do that for you;
• in a git repository, asks whether to put the Coders Talk git hooks into it (no, unless they are there already; see Git hooks).

coders-talk enable --yes goes ahead without asking, with --agent=claude-code,codex, --auto=off|on|team|push, --mcp=remove|keep and --git-hooks, --no-git-hooks or --no-trailers for the answers. coders-talk status shows the file, the sign-in, each agent's Coders Talk plugins and auto mode, and the git hooks of the repository you are in. coders-talk disable uninstalls the plugin and its marketplace from both agents and takes the git hooks out of every repository enable put them in; the sign-in, your settings and the file stay. coders-talk update lays the plugin out again in the new version and updates it in the agents, so the plugin's version is always the file's.

To send a session without going back into it, run coders-talk sessions in the folder it ran in: it lists that folder's Claude Code and Codex sessions (also those run at the top of its repository), newest first, with how many prompts each has, the start of the first one and whether it was sent already. coders-talk build 2 then shows the same summary and privacy check as /coders-talk:build, asks [y/N] and sends; coders-talk build alone takes the newest session, and a session id works too. It takes the same options (--private, --team=<slug>, --keep=<numbers>, --continues=<link>). It runs only in a terminal: neither an agent nor a script can answer its question, and there is no --yes. From a script, coders-talk preview <id> and coders-talk send <id> (with --agent=codex for Codex) do the two steps.

A Coders Talk connector your organisation added in claude.ai is not visible on your computer, so enable cannot find it. If the agent then lists the library's tools twice, turn one of the two servers off in /mcp.

From the agent

1. In Claude Code:

   Code

   Copy

   /plugin marketplace add coders-talk/coders-talk-plugin
   /plugin install coders-talk@coders-talk
   

2. Connect it to your account: /coders-talk:login. It opens coders.talk in the browser; check that the page shows the same code as Claude Code and press Connect. The plugin receives its token directly from the site and keeps it in ~/.coders-talk/credentials.json, readable by you only. There is no token to copy, and it never passes through the chat. The sign-in request names only the program asking (Claude Code, Codex or Coders Talk CLI) and its version, not your computer; the site lists the token under that name in Settings → Agent plugins.

Needs Node.js 20 or newer, nothing else.

Codex

codex plugin marketplace add coders-talk/coders-talk-plugin
codex plugin add coders-talk@coders-talk

Start a new session, then $coders-talk:login and $coders-talk:build (the same commands as below, with $ instead of /). For the library, also run codex mcp login coders-talk once in a terminal. Codex asks to run the plugin's command outside its sandbox: it needs the network to reach coders.talk and your home folder for the sign-in. The session is found through CODEX_THREAD_ID in ~/.codex/sessions (or CODEX_HOME), and HEAD at its start comes from the rollout's session_meta, so Codex needs no hook for it. Codex runs the plugin's hooks only for auto mode (see Auto mode), and only once you trust them in /hooks.

One repository serves both agents: Claude Code reads .claude-plugin/ and skills/, Codex reads .codex-plugin/, .agents/plugins/marketplace.json, codex/skills/ and codex/hooks.json (its manifest points there, so Codex never picks up hooks/hooks.json; the same scripts run with --agent=codex). Codex does not expand ${CLAUDE_PLUGIN_ROOT} or ${CLAUDE_SESSION_ID} in skills, so its skills give the script path relative to the skill file and pass --agent=codex.

Use

Sending the same session again updates its draft until you publish it.

Two rules:

• Commands that send your session (build and auto, and login and logout with them) run only when you type them. The agent never invokes them on its own.
• Putting a playbook into your repository (use) runs only when you type it too. The library may tell the agent that a session has a playbook; the agent can only pass that on to you.
• Searching the library is something the agent does by itself when a task calls for it (the lookup skill and the coders-talk MCP server). It sends a short description of the task and the stack, never the session. Switch it off in /mcp (Claude Code) or in ~/.codex/config.toml (Codex); see The library in your agent.