What is pn-best-of-n?

A tournament process that runs two or three competing implementations of the same specification in separate workspaces, verifies them, and compares the survivors.

Use it for algorithm, API, or refactoring choices with clear acceptance tests, but not for authentication, payments, secrets, or other security-sensitive paths.

It provides an objective way to choose between uncertain implementation approaches when tests or other checks can judge the results.

pn-best-of-n — Skill for Cursor by perniemann

Install

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.

Any agent

npx skills add perniemann/pnCore --skill pn-best-of-n

Clone the repo

git clone --depth 1 https://github.com/perniemann/pnCore

Made for: Cursor.

Or install pn-core, the plugin that ships this one along with the rest of its 133 skills, 19 commands, 9 agents, 1 MCP server.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/perniemann/pncore/pn-best-of-n/github.svg)](https://agentmods.dev/skills/perniemann/pncore/pn-best-of-n)

Your own site

<a href="https://agentmods.dev/skills/perniemann/pncore/pn-best-of-n"><img src="https://agentmods.dev/badge/skills/perniemann/pncore/pn-best-of-n/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/perniemann/pncore/pn-best-of-n"><img src="https://agentmods.dev/badge/skills/perniemann/pncore/pn-best-of-n.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 54 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,518 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00054	$0.01518
Opus 5	$0.00027	$0.00759
Sonnet 5	$0.00011	$0.00304
Haiku 4.5	$0.00005	$0.00152

Measured 8d ago against content hash d8c043a30512, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

pn-best-of-n 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 8d 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.

packages/pn-core-mcp/content/skills/orchestration/pn-best-of-n/SKILL.md · 121 lines

How it starts

The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Best-of-N (implementation tournament)

When to use

Ambiguous implementation choice with objective verification (tests, lint, typecheck)
Repeated skeptic failure on the same slice (same approach tried twice)
User sets useBestOfN: true or invokes /pn-best-of-n
Refactor with strong test harness, algorithm choice, API module shape after design exploration

When NOT to use

Auth, RLS, payments, secrets — use parallel review panel (pn-build-gate + subagent-routing.md), not competing implementations
Single-file edits, copy tweaks, shared mutable state without worktree isolation
No verification commands — run tests first or use pn-design-variants (design-only explore)
Full full_dev parallel specialists — use specialist decomposition instead

Design-only explore: use pn-design-variants first; this skill implements the winning shape.

Prerequisites

Written spec (plan slice, issue, or user brief) with acceptance criteria
Verification commands listed upfront (e.g. npm test -- <scope>, npm run lint)
Git worktree support (or best-of-n-runner subagent handles isolation)

Workflow

1. Confirm scope and gates

Restate spec in ≤5 sentences
List objective gates (commands + expected exit 0) — judges must not score candidates that fail these
Set N = 2 or 3 (default 2; use 3 only when ambiguity is high and budget allows)
Confirm scope is not security-critical; if it is, stop and route to review panel

Log report_usage when MCP available (pilot metric for P1 exit criteria).

MCP workflow (P2): when bestOfN.enabled is true in pn-core://config/features.json, use workflow_step('implementation_tournament', 0, {}) instead of ad-hoc fan-out. Skill-only path remains valid when the flag is off.

2. Fan-out — parallel builders

Spawn N Task subagents (subagent_type: best-of-n-runner) in parallel, each in an isolated worktree on the same spec with a different constraint:

Read the full file on GitHub · 121 lines

Changes

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.

8d ago First seen · 121 lines · 54 tokens per session scan A d8c043a30512

Subscribe to this mod's changes

pn-best-of-n is a skill published in the GitHub repository perniemann/pnCore (0 stars, last pushed 5d ago), licensed MIT. It adds 54 tokens to every session and 1,518 once invoked, about $0.0003 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 2026-09-03.

Code review LLM Testing TypeScript agents cursor mcp plugin

review

Enter the Review phase of CocoBrew. Aggregates findings from Code Quality Advisor, CocoCupper intelligence, spec compliance check, and decision coverage gate. Produces review.md with decision points. Requires developer approval before $ship can proceed.

Snowflake-Labs/cocoplus · 50 tokens

ork-verify

Verify that existing work is ready to merge, release, or hand off using an explicit evidence contract. Use when a request asks to verify, validate, prove, check readiness, run the relevant tests, or distinguish a claimed result from an observed one. Do not use to write missing tests or fix failures.

yonatangross/orchestkit · 65 tokens

pn-best-of-n