Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/seeincodes/skyfi-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/rules/seeincodes/skyfi-mcp/tech-stack-lock)<a href="https://agentmods.dev/rules/seeincodes/skyfi-mcp/tech-stack-lock"><img src="https://agentmods.dev/badge/rules/seeincodes/skyfi-mcp/tech-stack-lock/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/seeincodes/skyfi-mcp/tech-stack-lock"><img src="https://agentmods.dev/badge/rules/seeincodes/skyfi-mcp/tech-stack-lock.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00566 | $0.00566 |
| Opus 5.5 | $0.00226 | $0.00226 |
| Sonnet 5.5 | $0.00113 | $0.00113 |
| Haiku 4.5 | $0.00057 | $0.00057 |
Grade A, and why
tech-stack-lock scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Locked Technology Stack
These technology decisions are final. Do not switch to alternatives without explicit user approval.
Server Language & Runtime
- LOCKED: TypeScript 5.x on Node.js 20 LTS (local) / Cloudflare Workers (remote)
- DO NOT switch to Python, Go, Rust, or any other language for the MCP server
- DO NOT switch to Deno or Bun as primary runtime
- The demo agent IS Python (Google ADK) — that is a separate concern
MCP SDK
- LOCKED: @modelcontextprotocol/sdk (official TypeScript SDK)
- DO NOT use community MCP libraries or roll a custom protocol implementation
HTTP Framework
- LOCKED: Hono 4.x
- DO NOT switch to Express, Fastify, Koa, or other HTTP frameworks
- Hono was chosen for Cloudflare Workers native compatibility
Validation
- LOCKED: Zod 3.x for input validation and JSON Schema derivation
- DO NOT switch to ajv, joi, yup, or io-ts
- Zod schemas are the single source of truth for both validation and MCP inputSchema
Testing
- LOCKED: Vitest
- DO NOT switch to Jest, Mocha, or other test runners
Build
- LOCKED: tsup for TypeScript bundling
- DO NOT switch to esbuild directly, webpack, or rollup
Deployment
- LOCKED: Cloudflare Workers (primary remote) + Docker (self-hosted option)
- DO NOT switch to AWS Lambda, Vercel Functions, or other serverless platforms as primary
- Docker is the self-hosted escape hatch, not the primary deployment
Geocoding
- LOCKED: Nominatim (OSM) for geocoding
- DO NOT switch to Google Maps Geocoding, Mapbox, or other commercial providers as default
- Commercial fallback may be documented but Nominatim is the default
Observability
- LOCKED: Cloudflare Analytics + Logpush (cloud), Prometheus + Grafana (self-hosted)
- DO NOT add Datadog, New Relic, or other commercial APM as default
Adding New Dependencies
- New dependencies must be justified in a PR description
- Prefer dependencies that work on both Node.js and Cloudflare Workers
- Check Workers compatibility before adding any Node.js-specific package
- Keep the dependency count minimal — this is infrastructure, not an application
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 58 lines · 566 tokens per session scan A 1369c98d9a06
tech-stack-lock is a cursor rule published in the GitHub repository seeincodes/skyfi-mcp (0 stars, last pushed 7mo ago), licensed MIT. It adds 566 tokens to every session, about $0.0023 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other cursor rules, from other repositories
docs
All documentation files use MDX format with a specific structure.
turborepo
Turbo task-graph rationale, query graph primitives, and signed remote-cache key provisioning.
previews
Cloudflare Worker Previews pipeline - credential-free Preview build then cf previews deploy, cf previews delete, Wrangler-only secrets, Access before arming, agent loop.
react
React 19 SPA - client-only, layout and #/ imports, no fetch in useEffect, Compiler memo policy, providers, splitting, React Doctor.
browser-extension
Browser extension patterns and development practices.
simplicity
Simplicity - smallest correct diff; reuse, platform, installed deps before new code; no speculative abstraction.