Skip to content
Get Started KurrentDB

What is Capacitor

Pair programming used to mean two engineers sharing context as they worked. Now it’s an engineer and an agent — and the agent’s reasoning disappears the moment the chat scrolls. The PR ships; the why doesn’t. Reviewers see a diff with no co-author to ask, teammates inherit code nobody can explain, and future-you reopens the file in six months with a fresh mystery.

Capacitor isn’t observability for your agents, and it isn’t a transcript archive. It restores the shared context that disappeared when pair-programming turned solo — so a reviewer, a teammate, or future-you can ask “why did we pick this?” and get a grounded answer from what actually happened.

Concretely: it records every coding-agent session your team runs — Claude Code, Codex, Gemini CLI, Pi, OpenCode, Antigravity, GitHub Copilot CLI, Cursor, and Kiro — and surfaces them in a real-time dashboard, with full transcripts queryable from the CLI, the web UI, and (via MCP) from inside MCP-capable agents.

This page is the lens for everything else in the docs. If you’re trying to decide whether Capacitor is right for you, start at the Capacitor home page instead.

The two pieces

Section titled “The two pieces”

Capacitor has exactly two moving parts you interact with:

  • Your Capacitor server — a Kurrent-hosted Capacitor instance at https://<your-github-org>.kcap.ai. It stores your events, projects them into the dashboard, and serves the web UI. Kurrent provisions and operates it; you never deploy server infrastructure yourself.
  • The CLI — npm install -g @kurrent/kcap. It installs hooks into Claude Code, Codex, Gemini CLI, Pi, OpenCode, Antigravity, GitHub Copilot CLI, Cursor, and Kiro, streams transcripts to your Capacitor server in real time, and provides the commands you’ll use day to day (kcap recap, kcap review, kcap eval, …).

The CLI is open source. The server is not.

What gets recorded

Section titled “What gets recorded”

Once the CLI is set up, every coding-agent session is captured automatically:

  • Session lifecycle — start, end, interruptions, context compaction
  • Full transcripts — streamed to your Capacitor server with sub-second latency, so a teammate watching the dashboard sees turns appear as the agent emits them
  • Subagent trees — every spawned subagent with its own transcript and its own token tally
  • Tool calls — every Bash, Read, Edit, Write, etc. with timing, results, and the exact diff applied
  • Thinking blocks — the agent’s reasoning, captured alongside the turn that produced it
  • Token consumption — input, output, cache reads, and cache writes, per turn and per subagent
  • Repository context — git repo, branch, and PR linkage

You don’t run separate “record” commands; the hooks fire automatically once they’re installed.

What you do with it

Section titled “What you do with it”

The main surfaces:

  • The dashboard — https://<your-org>.kcap.ai. It opens on a personalized Home that surfaces what needs you — agents waiting, work to continue, fresh outcomes — and lets you organize repos into projects. Browse sessions and agents, watch them live, see ownership and sharing.
  • Outcome-first sessions — an ended session opens on an outcome view — what got done, the plan behind it, and what to try next — instead of the raw transcript, which stays a click away.
  • Recap, recall, and memory — kcap recap and the kcap-sessions MCP server replay and search past sessions (“have we worked on this before?”); team memory (kcap-memory) gives agents durable, shared knowledge that outlives any one session. Run from your terminal, or invoke as skills/MCP tools inside the agent.
  • Evaluations — score every session against safety, plan adherence, quality, and efficiency using LLM-as-judge. Findings flow back into per-repo clusters that admins curate and re-inject at the top of the next session, automatically. One session’s mistakes become the next session’s guardrails. The most distinctive thing Capacitor does — see Evaluations.
  • PR review — kcap review <pr-url> answers questions about a pull request grounded in the implementation transcripts behind it, not just the diff.
  • Hosted agents — launch a coding agent (any supported vendor) in an isolated worktree directly from the dashboard, controlled by a daemon on your machine. Multiple teammates can drive the same hosted session.
  • Review flows — have one agent drive an automated review loop against a hosted reviewer: submit a spec or diff, get findings, fix, re-review until it signs off. Generative review of work in progress — see Review flows.
  • Work items — Capacitor groups sessions around the issue, PR, and commit references they contain, so a whole line of work — the ticket, its PR, and every session that served it — sits in one place. Correlated automatically, or declared from inside a session. See Work items.
  • Analytics — ask questions about your sessions in natural language and get tables and charts back — cost, activity, PRs and commits, agent-vs-human attribution. See Analytics.

Where to go next

Section titled “Where to go next”