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.
npx skills add stdin/buy-vs-build --skill buy-vs-build-right-toolgit clone --depth 1 https://github.com/stdin/buy-vs-buildWrote 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/skills/stdin/buy-vs-build/buy-vs-build-right-tool)<a href="https://agentmods.dev/skills/stdin/buy-vs-build/buy-vs-build-right-tool"><img src="https://agentmods.dev/badge/skills/stdin/buy-vs-build/buy-vs-build-right-tool/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/skills/stdin/buy-vs-build/buy-vs-build-right-tool"><img src="https://agentmods.dev/badge/skills/stdin/buy-vs-build/buy-vs-build-right-tool.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 | $0.00075 | $0.01686 |
| Opus 5 | $0.00037 | $0.00843 |
| Sonnet 5 | $0.00015 | $0.00337 |
| Haiku 4.5 | $0.00007 | $0.00169 |
Grade A, and why
buy-vs-build-right-tool 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 11d 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Buy vs Build — Pick the Right Option, Not the Obvious One
The buy-vs-build ladder picks an ownership level; it does not pick which option at that level. When several options satisfy the rung, the flashy, popular, or familiar one is often wrong: it adds capability you never use and ownership you keep paying for. This is the reference for that second decision.
Reach for this when the choice is non-obvious or a one-way door. When it is obvious, just decide and keep moving.
Find the distinguishing requirement first
Name the one property that actually separates the options, then pick the option that fits it — nothing more:
- Directionality — one-way vs bidirectional; push vs pull; who initiates.
- Volume & latency — events per second, payload size, tail-latency budget.
- Consistency & ordering — transactional vs eventual; exactly-once vs at-least-once; ordered vs not.
- Durability & delivery — can a message be dropped, or must it survive a crash and retry?
- Failure mode — reconnect, backpressure, replay, idempotency, partial failure.
- Operational fit — what your runtime, proxies, clients, and infra already support and can debug.
If two options tie on the distinguishing requirement, pick the one with less long-term ownership (less infra to run, less API surface, easier to remove).
Decision tables
The middle column is the common over-reach. The "Fitting default" is the lower-ownership option that usually wins. Use the heavier option only when its extra capability is actually exercised — the right-hand column says when.
Realtime / transport (server↔client)
| You need | Common over-reach | Fitting default | Use the heavier option when |
|---|---|---|---|
| Server→client stream: feeds, notifications, progress, token streaming | WebSockets | Server-Sent Events (plain HTTP, auto-reconnect) | The client must also send a steady high-rate stream back — true full-duplex. |
| Bidirectional, low-latency: chat, collaboration, multiplayer, live cursors | Polling / SSE | WebSockets | (this is the case the heavier option is for) |
| Occasional client→server nudges | WebSockets | fetch / HTTP request |
You measured per-request overhead and it actually hurts. |
| Browser ↔ browser media/data, low latency | WebSockets relay | WebRTC | Peer-to-peer NAT traversal and media are the requirement. |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 11d ago First seen · 89 lines · 75 tokens per session scan A 18c2c40e7857
buy-vs-build-right-tool is a skill published in the GitHub repository stdin/buy-vs-build (3 stars, last pushed 2mo ago), licensed MIT. It adds 75 tokens to every session and 1,686 once invoked, about $0.0004 per session on Opus 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 skills, from other repositories
cr-next-type-llm
Cursor rules for next-type-llm.
implementation-loop
Disciplined plan-build-verify loop for implementing a feature or change. Use when building something new or modifying existing behavior, especially across multiple files.
debugging-loop
Systematic reproduce-isolate-fix-verify loop for bugs, test failures, and unexpected behavior. Use when something is broken and the cause is not yet known.
tw-migrate
Analyze a Tailwind CSS v3 project and generate a complete migration plan to v4. Scans config, CSS, and template files for v3 patterns and produces a prioritized checklist with exact find-and-replace commands.
tw-component
Generate a UI component using correct Tailwind CSS v4 utility classes, CSS variable theming, and accessibility best practices. Prevents v3 syntax hallucination by grounding output in verified v4 patterns.
docstruct
Documentation architecture discipline for AI agents. Use when creating, updating, auditing, or organizing software project documentation so docs stay clear, non-duplicated, source-of-truth driven, and token-efficient.