crewkitEarly Access
MarketplaceFeaturesPricingDocsGitHub
Sign InRequest Access
Works with Claude Code · MCP server included

The operating system for AI‑assisted engineering

Your team ships with AI agents every day — and every decision behind the code evaporates when the session ends. The meeting, the Slack thread, the reason you chose this pattern. crewkit remembers it, injects it, and makes it searchable, so every session starts where the last one left off.

For engineering leaders: an auditable trail of every convention accepted or overridden, and cost and quality attributed to the exact agent version that ran — AI-assisted work as a managed practice, not a black box.

Early access is free — no credit card while we onboard teams in batches.

Already have access? Sign in

terminal
$ crewkit code
crewkit v0.2.4
ready acme-corp/customer-portal
synced 5 agents, 6 skills, 3 playbooks (12 conventions)
launching main*
Remember
PRDs, transcripts, Slack threads, and past sessions become searchable project memory, injected automatically when a session starts
Govern
Playbooks and conventions with an auditable challenge log, role-based autonomy, and privacy-first detectors: no code egress, no LLM in the loop
Collaborate
Agents get a Slack handle, teammates see who's working on what live, and any session is one share link away

From one command to a team that remembers

Install, connect your context, and every session starts where the last one left off — with the evidence to improve what runs next.

01

Install in one command

brew, npm, choco, or curl. Run crewkit code: sign in inline, and your org and project resolve from the git remote.

02

Connect your context

Upload PRDs, transcripts, and contracts; import from Google Drive; capture Slack threads as project artifacts. Decisions stop evaporating.

03

Code with memory

crewkit code syncs your team's agents, skills, and playbooks, then injects the project context that matters. Mid-session, search everything the project knows — including prior sessions.

04

Improve on evidence

Every session records the exact version of every agent and skill that ran. Compare versions on cost, tokens, and quality scores; run A/B experiments with p-values; roll back what regressed.

Platform Features

Nothing your team decides gets lost

Decisions, docs, and past sessions become searchable project memory — with enforceable conventions, agents that work like teammates, and cost and quality attributed to the exact agent version that ran.

Team Memory That Outlives the Session

Institutional knowledge shouldn't live in one person's head. Upload PRDs, transcripts, contracts, and design docs; every analyzed conversation is indexed too. It all becomes searchable project memory — injected automatically when a session starts, so the agent already knows what your team knows.

  • Upload anything— PRDs, transcripts, contracts, design docs, and API specifications, all indexed for semantic search
  • Drive import and Slack capture— Pull from Google Drive or capture Slack threads as artifacts, human-confirmed before they are ever injected
  • Automatic context injection— Relevant artifacts load into every session at start, matched by project and topic
dashboard -- project memory
customer-portal | Context
auth-prd.md PRD injected
api-contract-v3.yaml Contract injected
2026-06-12-arch-review Transcript injected
#payments Slack thread Slack confirmed
Search: "why do we rotate refresh tokens?"
auth-prd.md -- section 4: token rotation
session "rotate refresh tokens" (May 12)

Agents That Work Like Teammates

Give an agent a Slack handle scoped to your project. @mention it in a thread and it replies in-thread, with full project context. Teammates see each other's active work, so nobody steps on anyone's toes.

  • Slack-native agents— @mention a project-scoped agent and it answers in-thread with everything the project knows
  • Active-work visibility— See teammates' open conversations and claimed blueprint tasks, live
  • Session sharing— Share any session with a link, control visibility and expiry, and annotate inline
slack -- #eng-checkout
maya 10:02
@checkout-agent why optimistic locking on order
rows? I thought we discussed row locks.
checkout-agent APP 10:03
Decided in the Q2 planning call -- optimistic
locking avoids contention on hot order rows.
Source: "Q2 planning transcript" (artifact).

Governance That Scales With Your Team

Enforce coding standards and best practices automatically. Junior developers get coaching mode, seniors get full autonomy — all controlled through role-based agent configuration.

  • Role-based behavior— Coaching, collaborative, and autonomous modes adapt agent output per developer
  • Playbook enforcement— Conventions auto-inject into every session, no manual setup per project
  • Convention tracking— Monitor adherence and overrides across your organization
.claude/agents/rails-expert.md
# Rails Expert
## Platform Guidelines (c714f6d)
You are a Rails 8 specialist...
## Organization Guidelines (acme-corp)
- Use RSpec (not MiniTest)
- Follow Rubocop Shopify guide
- Service objects for business logic
## Project Guidelines (customer-portal)
- PostgreSQL with pg_search
- Hotwire + Stimulus (no React)
Role: Junior Developer -> Coaching Mode

A/B Test Agent Configurations Like You Test Code

Stop guessing which prompts work best. Run controlled experiments, measure real outcomes, and deploy winning configurations with statistical confidence.

  • Controlled experiments— Split traffic between resource versions with automatic assignment
  • Real-time metrics— Track quality scores, cost, and token usage live
  • Statistical significance— Deploy only when p-values confirm the winner
dashboard -- experiment results
Experiment: swift-amber-falcon
Resource: rails-expert (agent)
Status: Running | Created: 3 days ago
Control Variant Delta
Sessions 47 53 +6
Quality 3.6/5 4.2/5 +0.6
Avg Cost $0.034 $0.029 -14.7%
Avg Turns 8.3 6.1 -26.5%
Statistically significant (p=0.023)
Recommendation: Deploy variant

Full Visibility Into Every AI Session

Track sessions, costs, and outcomes across your entire organization. Drill into any conversation to see exactly what happened, what it cost, and how it performed.

  • Session telemetry— Turns, tokens, costs, tool usage, and duration per conversation
  • Team analytics— Timeseries, per-agent, per-project, and cost-breakdown dashboards
  • AI-powered analysis— Automatic session summaries, coaching tips, and quality scores
dashboard -- analytics overview
acme-corp | Last 7 days
Sessions
147 total | 3.1% revert rate
Avg duration: 12m 34s | Avg turns: 8.2
Cost
$18.42 total | $0.125 avg/session
1.2M input tokens | 89K output tokens
By Agent
rails-expert 52 sessions quality 4.2/5
frontend-expert 38 sessions quality 3.9/5
api-designer 31 sessions quality 4.4/5

Built for the full AI development lifecycle

Prior Work, Searchable

Every analyzed conversation is indexed, so "has anyone touched this before?" returns the session that did the work — privacy-gated for sensitive and private sessions.

Playbook-Driven Consistency

Define conventions once at the org level. They propagate to every project and every developer — no drift, and every exception logged.

Continuous Improvement Loop

Configure resources, sync to your team, measure performance, and iterate. Your agents get better every sprint.

Resource Marketplace

Pre-built agents, skills, and playbooks

Browse a growing library of ready-to-use resources. Pick what fits your stack, customize for your team, and sync to every developer. Every resource is versioned, measured, and rollback-able.

Agents

Specialized AI personas with deep domain expertise

rails-expert
Rails conventions, migrations, testing
frontend-expert
React, TypeScript, accessibility
api-designer
REST/GraphQL API design patterns
security-reviewer
OWASP, auth, vulnerability scanning
and more in the marketplace

Skills

Reusable capabilities that extend what agents can do

create-pr
Draft PRs with proper descriptions
review-code
Structured code review with checklist
run-tests
Execute and analyze test suites
plan-feature
Break features into tasks and subtasks
and more in the marketplace

Playbooks

Org-wide conventions and coding standards, enforced automatically

Rails 8 Standards
Rubocop, RSpec, service objects
React + TypeScript
Strict mode, hooks, testing library
API Security
Auth, rate limiting, input validation
CI/CD Pipeline
GitHub Actions, deploy checks, rollbacks
and more in the marketplace

Your knowledge compounds — privately

Every convention challenged, session analyzed, and version measured makes your team's setup sharper. Nothing leaves your organization: your memory works for your team, and only your team.

Platform— curated base resources
Organization— team standards & conventions
Project— repo-specific context
Versioned
Every change, rollback-able
Measured
Cost & quality per version
Request Early Access

Your team already has the knowledge. Keep it.

One command to install, one command to start. Early access is free — request your invite and it lands by email.

Request Early AccessTalk to Sales
crewkit

crewkit is the institutional memory and governance layer for AI-assisted engineering: agents work as teammates, prior work is one search away, and every session is attributed to the exact agent version that ran it.

Currently in early access.

Product

  • Marketplace
  • Features
  • Pricing
  • Documentation

Resources

  • GitHub
  • Changelog
  • Report Issue
  • System Status

Company

  • About
  • Privacy
  • Terms

© 2026 Karibew Inc. All rights reserved.

Command Palette

Search conversations, projects, playbooks, and more