Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/DataLab-atom/EvoAnynpx agentmods add skills/datalab-atom/evoany/huntWrote 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/datalab-atom/evoany/hunt)<a href="https://agentmods.dev/skills/datalab-atom/evoany/hunt"><img src="https://agentmods.dev/badge/skills/datalab-atom/evoany/hunt/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/datalab-atom/evoany/hunt"><img src="https://agentmods.dev/badge/skills/datalab-atom/evoany/hunt.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 | $0.00024 | $0.00892 |
| Opus 5.5 | $0.00010 | $0.00357 |
| Sonnet 5.5 | $0.00005 | $0.00178 |
| Haiku 4.5 | $0.00002 | $0.00089 |
Grade A, and why
hunt 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 2mo 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 — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/hunt — Find & Deploy a Codebase
Usage: /hunt <task description>
Example: /hunt I want SOTA on CIFAR-100-LT
Step 1: Literature & Repo Search
Run these sources in parallel for the best coverage:
Source A — Papers With Code (via browser)
browser navigate: https://paperswithcode.com/sota/<relevant-benchmark>
Or search:
browser navigate: https://paperswithcode.com/search?q_type=&query=<keywords>
Extract: SOTA methods, their paper titles, official code links.
Source B — arXiv (if arxiv-watcher skill is installed)
/arxiv-watcher <keywords>
Returns structured list of recent papers with abstracts and repo links. Use when the task involves a specific ML problem (image classification, NLP, etc.).
Source C — GitHub search
exec: gh search repos "<keywords from task>" --sort stars --limit 20 \
--json name,url,description,stargazersCount,updatedAt
Try keyword variations:
- Core method name: e.g. "CIFAR-100 long-tail"
- Algorithm name: e.g. "balanced softmax" "decoupled training"
- Task type: e.g. "imbalanced classification pytorch"
Source D — Summarize papers quickly (if summarize is installed)
For the top candidate papers, get their key contributions fast:
/summarize <arxiv_pdf_url>
Or for README of candidate repos:
/summarize <github_repo_url>
Step 2: Evaluate Candidates
After collecting results from all sources, pick top 3–5 candidates. For each, check:
- Stars / recency / last commit date
- Has eval script or benchmark command?
- Clear setup instructions?
- License allows modification?
Present to user:
Found 3 candidates:
1. ⭐ 2.3k user/balanced-meta-softmax — BALMS, ECCV 2020, last commit 3mo ago
2. ⭐ 1.8k user/long-tail-recognition — Multiple methods, active maintenance
3. ⭐ 950 user/cifar-lt-baseline — Clean PyTorch baseline, good eval script
Recommend #1. Proceed? (or pick another)
Wait for user confirmation before proceeding.
Step 3: Clone and Set Up
exec("git clone <repo_url> ~/evo-workspace/<repo_name>")
exec("cd ~/evo-workspace/<repo_name> && cat README.md")
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.
- 2mo ago First seen · 139 lines · 24 tokens per session scan A 54340e700804
hunt is a skill published in the GitHub repository DataLab-atom/EvoAny (37 stars, last pushed 6mo ago), licensed Apache-2.0. It adds 24 tokens to every session and 892 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 2026-08-30.
hunt compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| huntDataLab-atom/EvoAny | Skill | 37 | 6mo ago | Apache-2.0 | A | 24 | Claude Code, Codex |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |