AI analysis grounded in the code graph — computed facts, not vibes · 2026-10-05T03:03:24Z
Paperclip is a Node.js server with a React (TypeScript) UI for orchestrating multiple AI coding/business agents as if they were employees. It provides an "org chart" abstraction (CEO, CTO, engineers, marketers) over heterogeneous agent runtimes — OpenClaw, Claude Code, Codex, Cursor, or plain HTTP/bash processes — surfaced through pages such as Inbox, PipelineSettings, Costs, CompanySkills, and IssueDetail. It's aimed at teams or individuals running many concurrent autonomous agents who need a dashboard for goals, budgets, task tracking, and cost governance rather than raw terminal output.
The 198-star gain sits within a very active release and commit cadence: five tagged releases between 18 June and 22 July 2026, and 30+ commits in the sampled window touching release infrastructure (feat(release): add human-gated beta channel, feat(release): add smoke-gated nightly channel and lane-separated Docker tags), import/export of whole companies, and multi-provider agent connections. This suggests the growth tracks genuine feature velocity — particularly around company import/export, secrets handling, and multi-agent provider support — rather than a single marketing event; the README's "bring your own agent" framing aligns with commits like feat(apps): support multiple provider connections and feat(adapter-codex-local): add secure device-login building blocks. No churn or viral-post evidence is given, so the causal link to this specific day's spike can't be confirmed from the facts alone.
What changed recently, how it's actually built (from the code graph), and whether you should care. Free account — no card, no spam.