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 EzraApple/skills-init --skill arenagit clone --depth 1 https://github.com/EzraApple/skills-initWrote 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/ezraapple/skills-init/arena)<a href="https://agentmods.dev/skills/ezraapple/skills-init/arena"><img src="https://agentmods.dev/badge/skills/ezraapple/skills-init/arena/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/ezraapple/skills-init/arena"><img src="https://agentmods.dev/badge/skills/ezraapple/skills-init/arena.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 | $0.00031 | $0.00569 |
| Opus 5.5 | $0.00012 | $0.00228 |
| Sonnet 5.5 | $0.00006 | $0.00114 |
| Haiku 4.5 | $0.00003 | $0.00057 |
Grade A, and why
arena 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 18d 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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Compare independent solutions
Have several fresh subagents solve the same task independently, then integrate one coherent result. Each candidate produces the whole artifact; splitting one implementation among workers is a different workflow.
Frame the task
Define the artifact, constraints, shared inputs, allowed changes, and checks that would establish success. Set a few concrete comparison criteria before reading the candidates. Choose a candidate count and resource budget appropriate to the decision; avoid multiplying attempts without a useful reason.
Give every candidate the same task brief and starting state, including relevant uncommitted changes. If the user specified alternatives to compare, preserve that comparison. Otherwise let the candidates choose their approaches independently.
Generate independently
Start each candidate in a fresh context window without inherited conversation history, the builder's preferred solution, or another candidate's reasoning. Share necessary facts and repository guidance through one concise grounding packet. Keep the judge's scoring notes outside candidate contexts.
Prepare a separate worktree or output directory for each candidate from that starting state. Candidates may write only their assigned outputs and run checks within their authorized scope; they must not edit the shared checkout or other candidates. External actions still require the user's authorization.
Run candidates in parallel when available. Require a complete artifact and a short rationale naming important tradeoffs. If one fails, report the missing candidate and assess whether the remaining comparison is still useful.
Compare and integrate
After candidates finish, start a fresh, read-only judge with the original task brief, constraints, relevant shared inputs, criteria, and completed artifacts. Use neutral labels when practical. Have it assess each criterion with evidence and recommend a base. Do not judge unfinished outputs.
The parent reads every candidate, checks the judge's evidence, and chooses the base. Resolve disagreements through actual artifacts and constraints; agreement between agents does not establish correctness. Prefer a maintainable boundary when valid candidates are otherwise close. If none meets the task constraints, explain the failure rather than forcing a winner.
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.
- 18d ago First seen · 65 lines · 31 tokens per session scan A d743aafec90d
arena is a skill published in the GitHub repository EzraApple/skills-init (2 stars, last pushed 19d ago), licensed MIT. It adds 31 tokens to every session and 569 once invoked, about $0.0001 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 2026-10-16.
Other skills, from other repositories
oma-scholar
Search academic literature and generate, validate, or compare Knows paper sidecars. Use for claim/evidence analysis and literature synthesis.
pkf-writing
Author, update, convert, validate, persist, and render PKF (.pkf) content through message-inline, Library Asset, or Memory Page carriers. Use for structured reports, knowledge pages, delivery cards, or any .pkf body. NOT for recall, memory-graph placement, or curation decisions — those belong to the memory skill.
monorepo-management
Master monorepo management with Turborepo, Nx, and pnpm workspaces to build efficient, scalable multi-package repositories with optimized builds and dependency management. Use when setting up monorepos, optimizing builds, or managing shared dependencies.
render-3d-product-showcase
Assemble a premium 3D product-showcase ad from a config — four beat clips (an orbiting hero rotation, a macro push-in, a physics reveal, a typographic close) normalized to the chosen backdrop colour, hard-concatenated in order, closed on a deterministic Playwright brand end card, and mixed under one instrumental bed…
render-airdrop-carousel
Assemble a viral iOS "AirDrop" notification-carousel video ad (≈6–8s, 9:16) from a brand line plus 6–16 real product photos — a native AirDrop share-sheet card ("Brand would like to share a · Decline / Accept") springs up and its preview window CYCLES through the products, landing on a range/lineup payoff with an…
email-lifecycle-flow-writer
Step of the email-lifecycle playbook: writes one lifecycle flow as a build sheet for the brand's email tool, every message finished in the brand voice. Run email-lifecycle instead; not for direct use.