npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

js-evolution-engine

v0.2.0

Published

OADA (Observe-Analyze-Decide-Act) autonomous evolution engine for AI agents. Three-pipeline framework with Issue-driven decisions, PR-based execution, and risk-aware verification.

Readme

js-evolution-engine

OADA (Observe → Analyze → Decide → Act → Verify) autonomous evolution engine for AI agents. A pluggable, host-agnostic framework with a three-pipeline architecture: Issue-driven decisions, queue-based execution, and risk-aware verification.

Node ESM License

What it is

js-evolution-engine orchestrates an autonomous self-evolution loop for AI-driven projects. It separates what to do (the intelligence pipeline) from how to do it (the execution pipeline) and whether the result is safe to keep (the verify pipeline). Each pipeline can be invoked independently from the CLI, run on a schedule, or composed programmatically.

┌──────────────────────┐    ┌──────────────────────┐    ┌──────────────────────┐
│  Intelligence        │    │  Execution           │    │  Verify              │
│  Observe → Analyze   │ →  │  Claim → Dispatch    │ →  │  Audit → Policy      │
│        → Decide      │    │       → Modify       │    │       → (Auto-)Merge │
│  Output: Issues +    │    │  Source: Queue or    │    │  Source: GitHub PRs  │
│         Queue        │    │           GitHub     │    │       or local       │
└──────────────────────┘    └──────────────────────┘    └──────────────────────┘
        ↑                            ↑                            ↑
        └────── HostContext ─────────┴────── AI Client ───────────┘
                  (logger, notifier, action handlers,
                   intelligence store, analytics, …)

5-minute quickstart

npm install js-evolution-engine

Create oada.config.mjs in your project root:

import { MockAIClient, ActionTypeRegistry, ActionTypeSpec } from 'js-evolution-engine';

const actionRegistry = new ActionTypeRegistry();
actionRegistry.register(new ActionTypeSpec({
  name: 'send_email',
  description: 'Send a notification email',
  promptHint: 'Send an email (params: to, subject, body)',
  defaultRisk: 'low',
}));

export default async function ({ cwd }) {
  return {
    aiClient: new MockAIClient({ /* or your own AI subclass */ }),
    actionRegistry,
    host: {
      basePath: cwd,
      appName: 'my-agent',
      logger: console,
      actionHandlers: {
        send_email: async (action, ctx) => {
          // your real send-email logic here
          ctx.logger?.info(`would email ${action.params?.to}`);
          return { success: true, message: 'sent' };
        },
      },
    },
  };
}

Then:

npx oada intel              # observe + analyze + decide → queue actions
npx oada exec  --limit 5    # consume the queue
npx oada decisions          # inspect queue state
npx oada verify --auto      # (github mode) audit PRs + auto-merge low-risk

See examples/minimal-demo (from a git checkout) or node_modules/js-evolution-engine/examples/minimal-demo after npm install — only source files are published; run the demo to create data/ locally.

For an example of injecting external markdown via agentContextDocs, see examples/cyber-taoist-demo in the repo (node_modules/js-evolution-engine/examples/cyber-taoist-demo/ after install — sources only), and Injecting Agent Context Documents in the host-adapter guide. On npmjs.com, use the repo copy: examples/cyber-taoist-demo tree and HOST_ADAPTER.md on GitHub.

Documentation

Public API

import {
  // Engine + pipelines
  EvolutionEngine, IntelligencePipeline, ExecutionPipeline, VerifyPipeline,
  // Host abstractions
  NULL_HOST, normalizeHost,
  // AI
  BaseAIClient, MockAIClient, AIError,
  PromptBuilder, promptBuilder,
  // Registries
  ACTION_REGISTRY, ActionTypeRegistry, ActionTypeSpec,
  OBSERVATION_REGISTRY, ObservationSourceRegistry, ObservationSourceSpec,
  // Queues
  DecisionQueue, FeatureRequestQueue,
  // Adapters
  HumanGuidanceReader, EvolutionLogger,
  // Helpers
  scanProjectStructure, QueryResolver, GoalProvider,
  // GitHub integration
  GitHubIssueManager, OADA_LABEL_DEFS,
  // Time helpers (vendored)
  isoBeijing, todayBeijing, nowBeijing, nowBeijingStr,
} from 'js-evolution-engine';

Status

v0.2.0 — adds agentContextDocs for host-injected authoritative markdown, optional ActionTypeSpec.layer, and ships docs/ + examples/ in the npm tarball. Core flows remain covered by unit tests; CLI and demos run end-to-end. Advanced features (e.g. detailed PR auto-merge policies, sub-agent dispatch via OpenClaw Gateway) are still simplified vs. the original; see CHANGELOG.md.

License

MIT — see LICENSE.