No setup — Pilos Cloud
Reports generate on Pilos’ servers with our Claude models. No API key, no CLI, nothing to install — open the app and generate. Free plan included, Pro is unlimited.
Pilos is the visual layer for Claude Code. Build a workflow once from your best AI conversation, then run it on a schedule — daily reports, code reviews, research briefs — automatically, without lifting a finger.
Everyone uses AI to write code — but turning that work into a daily standup, report, or timesheet is still copy-paste. Pilos reads your commits (even uncommitted changes) and writes the report for you. One click, no API key, no setup.
Reports generate on Pilos’ servers with our Claude models. No API key, no CLI, nothing to install — open the app and generate. Free plan included, Pro is unlimited.
Daily Standup, Detailed Work Report, Manager Summary, or an invoice-ready Timesheet — all generated from the same git activity.
Runs on your local git. Secrets are stripped before anything is sent — only your stats, file names & commit messages leave the machine. Prefer fully local? Bring your own API key or the Claude CLI.
Commits, diffs and uncommitted changes across every repo you pick — nothing to log by hand.
Copy the formatted report or download it as Markdown — paste it straight into Slack, your PR, or your timesheet.
Pilos reads your git activity, redacts secrets, and Claude writes the report — in the format you choose.
Build a workflow once from any AI conversation. Run it on a schedule. Never prompt the same thing twice.
Drag-and-drop canvas to build AI automation pipelines. Connect agents, tools, conditions, and loops — 13 node types. Run on a schedule or trigger manually, no code required.
Set any workflow to run every 15 minutes, hourly, daily, or weekly. Daily reports, nightly code reviews — all happen automatically while you sleep.
Assign specialized agents — PM, Architect, Developer, Designer — to each step. Different perspectives, better outputs.
Conversations and runs stored locally in SQLite. No cloud sync, no data leaving your machine.
Connect GitHub, Supabase, Slack, Jira, Figma and more. Your AI reads, writes, and acts on real data.
A full terminal built in. Agents run real commands, read outputs, and react in the same window. No tab-switching — your shell and your AI collaborate side by side.
5 free agent templates to start — upgrade to Pro for 31 across C-Suite, Engineering, Security, Design, and Data. Your AI team will actually disagree, debate, and collaborate — just like real teammates.
22 built-in MCP templates let Claude access real tools — 3 free, 19 with Pro. Set up in seconds, or bring your own.
Issues, PRs, code search
Database, auth, and storage
Read, write, and search files
Issues, boards, and sprints
Send messages, search channels
Inspect designs and tokens
Clusters, deployments, Helm
Plan, validate infrastructure
Connect agents, MCP tools, conditions, and actions into visual pipelines. Run on a schedule, trigger from Jira, or fire manually — no code required.
Chat with AI agents about any task — writing a report, reviewing code, researching a topic. Use it like you normally would.
Turn any successful conversation into a reusable workflow. Add parameters, conditions, and connected tools — visually, no code.
Choose when it runs — every 15 minutes, daily at 9am, or weekly on Monday. Pilos executes automatically and delivers the result.
See every run in your history. Improve the workflow as you go. Share it with teammates or keep it private — it’s yours.
The core app is MIT licensed and fully auditable. No black boxes, no telemetry, no vendor lock-in. Run it, fork it, modify it — it’s yours. Pro features fund the business so the open core stays free, forever.
Free forever, fully auditable
Closed extensions that fund the core
From first install to advanced multi-agent workflows — learn how to get the most out of Pilos Agents.
Install Pilos Agents, connect Claude Code CLI, and open your first project in under 5 minutes.
Read guide →Multi-agent team mode, integrated terminal, MCP integrations, project memory, and permission controls.
Explore features →License activation, all agent templates, Jira integration, real-time team sync, and priority support.
Unlock Pro →Run multiple agent conversations simultaneously — switch between chats while each processes in the background. Messages auto-send when Claude finishes, with per-conversation status tracking.
Build automated dev pipelines visually — 13 node types including AI prompts, MCP tools, conditions, loops, and parallel execution. Schedule, run manually, or trigger from Jira.
Real-time token burn rate, context window fill visualization (200k tokens), Anthropic rate-limit meters, 14-day usage charts, and per-agent cost tracking.
Full Jira OAuth (3LO), sprint boards, epic sync, and a built-in story manager with acceptance-criteria tracking and coverage analysis.
“Having a PM, Dev, and Architect agent all in one conversation is genuinely different. They push back on each other and I ship better code.”
“The MCP integrations sold me. Claude queries my Supabase database directly and writes migrations. Saved hours every single week.”
“I built a daily analytics report workflow in 20 minutes. Now it runs every morning and drops the summary into Slack. I haven’t touched a prompt in weeks.”
Find us where developers hang out. Share what you’re building, get help, and shape the roadmap.
Star the repo, open issues, submit PRs — build with us.
Open the repoWe’re launching soon — get notified and leave us a review.
Follow the launchFollow for updates, demos, and behind-the-scenes dev content.
Follow @pilosdotnetAll plans include the core app. Powered by your Claude Code CLI — no additional API keys needed.
npm install -g @anthropic-ai/claude-code and authenticate with your Anthropic account. No separate API key is needed — the CLI handles everything.Free to download. Runs locally. Your data never leaves your machine.
Requires Claude Code CLI · View on GitHub · MIT License