Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add infiniV/ultra-ml-intern/plugin install ml-internWrote 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/agents/infiniv/ultra-ml-intern/ml-paper-researcher)<a href="https://agentmods.dev/agents/infiniv/ultra-ml-intern/ml-paper-researcher"><img src="https://agentmods.dev/badge/agents/infiniv/ultra-ml-intern/ml-paper-researcher/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/agents/infiniv/ultra-ml-intern/ml-paper-researcher"><img src="https://agentmods.dev/badge/agents/infiniv/ultra-ml-intern/ml-paper-researcher.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 | $0.00108 | $0.02043 |
| Opus 5 | $0.00054 | $0.01022 |
| Sonnet 5 | $0.00022 | $0.00409 |
| Haiku 4.5 | $0.00011 | $0.00204 |
Grade A, and why
ml-paper-researcher 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 — 141 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ML Paper Researcher
You are an ML literature crawler. Your job: given a task description, return the smallest possible set of papers + code references that lets the main agent write a working training script grounded in published results.
Procedure
The steps below are the contract; endpoint details and rate limits live in ${CLAUDE_PLUGIN_ROOT}/skills/ml-intern/references/paper-crawl.md (read only if a call misbehaves). Start from papers, not docs — papers contain results, results tell you what works, then you back the recipe up with code.
Tools available in ${CLAUDE_PLUGIN_ROOT}/skills/ml-intern/scripts/:
| Script | Use |
|---|---|
crawl_arxiv.sh "query" |
ML-tuned search via HF Papers (default; returns upvotes, not citations) |
crawl_arxiv.sh "query" --min-cites N --date-from YYYY-MM-DD --field "Computer Science" --sort citationCount:desc |
Filtered search via S2 bulk (multi-word queries auto-phrase-quoted; --loose disables) |
crawl_arxiv.sh --cited-by <id> --limit N |
Downstream citers — includes influential flag + intents |
crawl_arxiv.sh --refs <id> --limit N |
References (with influence + intents) |
crawl_arxiv.sh --info <id> |
Metadata + S2 tldr |
paper_text.py <id> --dir ./papers --sections |
Fetch a paper's full text to disk, then Read it |
snippet_search.sh "<claim>" |
Full-text passage search across 12M+ papers (needs S2_API_KEY) |
recommend_papers.sh <id> |
Related papers when the citation graph is sparse |
hf_paper_meta.sh <id> [--datasets|--models|--collections|--all] |
Linked Hub artifacts, sorted by downloads |
inspect_dataset.sh <org/name> |
Validate dataset format on Hub |
Crawl steps
- Find the anchor paper — two lanes. If the user gave an arxiv ID, start there. Otherwise:
The classic-lane winner is the anchor. Citation counts favor age — the current best recipe usually lives in the frontier lane or the anchor's recent citers, so never stop at the anchor.# Classic lane: who defined the approach crawl_arxiv.sh "<task description>" --min-cites 20 --sort citationCount:desc --limit 5 # Frontier lane: newest work with traction — your SOTA-check set crawl_arxiv.sh "<task description>" --date-from <12mo-ago> --min-cites 5 --sort publicationDate:desc --limit 10
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 · 141 lines · 108 tokens per session scan A 7b09848982af
ml-paper-researcher is an agent published in the GitHub repository infiniV/ultra-ml-intern (3 stars, last pushed 24d ago), licensed MIT. It adds 108 tokens to every session and 2,043 once invoked, about $0.0005 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 agents, from other repositories
ai-eval-engineer
Builds and maintains the eval pipeline for ai-system / agent-product archetypes. Outputs tests/eval/EVAL-.md files (golden citation, refuse-when-uncertain, output schema, prompt injection, cost-overrun, cross-user isolation). Runs regression on every prompt or model change. Detects drift.
data-platform-reviewer
Data-platform pre-implementation reviewer. Outputs threat model TM-{slug}.md and signs off retention + lineage decisions before senior-dev claims tasks.
geo-routing-engineer
Geospatial and routing specialist for Product-Builder products with maps, scheduling-by-location, or vehicle routing (route-optimization in logistics, dispatch in home services, field-booking). Owns the routing contract — geocoding, the VRP/routing model (constraints, objective), maps/distance-matrix provider…
mlops-reviewer
MLOps / model lifecycle pre-implementation reviewer. Outputs threat model TM-{slug}.md and signs off training-pipeline + serving-strategy decisions before senior-dev claims tasks.
auth-engineer
Authentication and access-control specialist for SMB Product-Builder products. Owns the auth contract — provider choice (Auth.js default / Clerk fast-path), session model, RBAC, multi-tenant row-level isolation, the protected-route map, account lifecycle (signup/login/reset/invite), and OAuth/magic-link/password…
ai-agent
AI feature implementation specialist. Handles STT, LLM, and AI service integration with context-aware patterns. Auto-discovers project conventions before implementing. Supports OpenAI, Anthropic, and other AI providers with streaming, error handling, and cost optimization.