396
total weekly downloads
36 packages
@16ane/mcp-server
MCP server that gives an agent a budget-capped 16ane sandbox wallet: pay x402 services in tANE, a free test token with no cash value
@16ane/sdk
TypeScript SDK for the 16ane agent-payments sandbox: vault budgets, x402 payments and on-chain receipts in tANE, a free test token with no cash value
@bytesbrains/claude-code-cruise
BytesBrains Cruise for Claude Code: your key's models, budget and spend over MCP, a skill for lanes and refusals, a spend status line, and guided setup. The package root is the plugin root, for a marketplace's npm source.
@bytesbrains/cruise
Run Cruise — one OpenAI-compatible endpoint in front of every model provider, with budgets that stop — in Docker or on your own Cloudflare account.
@bytesbrains/cruise-worker
The Cruise Worker, built and ready to deploy: one OpenAI-compatible endpoint in front of every model provider, with budgets that stop.
@bytesbrains/llm-cost-control
LLM cost-control layer — response caching, budget gates, model routing, per-tenant metering. The margin-protector for managed AI runs.
@bytesbrains/maktub-sdk
TypeScript SDK for the Maktub Protocol — decentralized conditional execution engine on Base L2
@bytesbrains/mcp-textbrowser
MCP server — text-first headless browser with DOM + OCR. Zero image tokens by default. 5-15x cheaper than screenshot-based browser MCPs.
@bytesbrains/n8n-nodes-cruise
BytesBrains Cruise community nodes for n8n — OpenAI-compatible chat through Cruise keys, budgets, and lanes.
@bytesbrains/openclaw-cruise-provider
BytesBrains Cruise provider plugin for OpenClaw — live GET /v1/models discovery.
@bytesbrains/opencode-cruise
BytesBrains Cruise plugin for OpenCode — budgets, lanes, and OpenAI-compatible routing via the Cruise gateway.
@bytesbrains/pi-agent-supervisor
Runtime safety net for AI agents — blocks dangerous commands, protects files, enforces rate limits, and records sessions.
@bytesbrains/pi-auditor
Audit repo files for RAG digestibility — find god files, measure chunk fragmentation, and generate split suggestions.
@bytesbrains/pi-ci-gate
CI observability gate for AI agents — view workflow runs, job statuses, and logs from Gitea Actions with safety controls.
@bytesbrains/pi-contrib-gate
Contribution gateway for AI agents — enforce branch naming, conventional commits, pre-commit quality gates, and PR automation.
@bytesbrains/pi-docker-logs
Docker log gateway for pi agents — tail logs from Docker containers in the wrok.in AI Factory via the docker CLI.
@bytesbrains/pi-factory-gate
AI Factory gateway for pi agents — dispatch jobs, manage environments, list agents/workers/workflows, stream results from a wrok.in orchestrator.
@bytesbrains/pi-loki-gate
Loki log gateway for pi agents — query Grafana Loki for container/job/worker logs from the wrok.in AI Factory.
@bytesbrains/pi-project-gate
Project orchestration gate for AI agents — structured issues, WIP limits, dependency blocking, and auto-generated release notes.
@bytesbrains/pi-review-gate
CI-level merge guardian for AI agents — required reviewers, breaking change detection, stale PR cleanup, and review automation.
@bytesbrains/pi-telegram-bridge
Telegram bot bridge for pi agents — send messages, ask questions, and listen for human replies via Telegram.
@bytesbrains/pi-textbrowser
Headless browser for Pi — browse the web with DOM + OCR text maps. No image tokens, 10-50x cheaper than screenshot-based browsing.
@bytesbrains/pi-tool-awareness-gate
Tool limitation awareness layer for AI agents — intercepts tool results, infers quality signals, injects limitation reminders, and logs structured envelopes for evaluation.
@bytesbrains/weblocks
Block engine for AI-composable web apps — the AI composes a SiteManifest from a fixed block catalog; the engine validates and renders it to static HTML. Snap-together "Lego" web building blocks, shareable across repos.
maktub-protocol
Maktub Protocol — Decentralized conditional execution engine on Base L2