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 bravew/trove --skill trove-arenagit clone --depth 1 https://github.com/bravew/troveWrote 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/bravew/trove/trove-arena)<a href="https://agentmods.dev/skills/bravew/trove/trove-arena"><img src="https://agentmods.dev/badge/skills/bravew/trove/trove-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/bravew/trove/trove-arena"><img src="https://agentmods.dev/badge/skills/bravew/trove/trove-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 | $0.00066 | $0.00576 |
| Opus 5 | $0.00033 | $0.00288 |
| Sonnet 5 | $0.00013 | $0.00115 |
| Haiku 4.5 | $0.00007 | $0.00058 |
Grade A, and why
trove-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 10d 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 — 36 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Trove · v2026.7.4
Session Init
This skill ships Trove conventions. Prefer existing project patterns over generic best practices when they conflict.
If a sibling skill in this plugin matches the request more directly, defer to it. See AGENTS.md (or docs/routing.md in the marketplace) for the per-plugin routing index.
trove-arena
Fan out several independent attempts at the same hard problem, then synthesize the best of them. This is expensive — N parallel agents cost N times the tokens. Use it only when one attempt would lock in the wrong shape and the decision is hard to reverse. For ordinary non-trivial work, use trove-architect.
Phases
- Frame. State the problem and write the rubric (3-6 concrete, gradeable criteria) before spawning. The rubric is the picker's tool; candidates never see it.
- Fan out. Spawn N candidates on the same task. Each writes to its own isolated path — a separate worktree (
trove-worktree) or/tmpdir — so they never share state. Keep N small (2-4); more is rarely worth the cost. - Cross-judge. Read-only pass on a fresh context that scores each candidate against the rubric. Run it after candidates finish, not while they write (a judge reading mid-write sees empty output).
- Pick a base. The strongest candidate overall, by the rubric.
- Graft. Hand-port the best ideas from the losing candidates into the base. Record why each rejected candidate lost — the rejection notes are the highest-signal part of the record.
- Verify. Prove the grafted result works (
trove-verify/ its own tests).
Convergence vs divergence
If candidates converge on the same approach, that consensus is a strong signal — ship it. If they diverge sharply, that means the problem is underspecified: reframe it, don't average the candidates into mush.
When not to use
Mechanical changes, bug fixes, or anything with an obvious single approach. Sending those to the arena burns tokens for no gain — trove-architect or a direct change is correct.
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.
- 10d ago First seen · 36 lines · 66 tokens per session scan A 0c50b03aa138
trove-arena is a skill published in the GitHub repository bravew/trove (10 stars, last pushed 11d ago), licensed MIT. It adds 66 tokens to every session and 576 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…