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 evoelsewhere/evoflux --skill super-researchgit clone --depth 1 https://github.com/evoelsewhere/evofluxWrote 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/evoelsewhere/evoflux/super-research)<a href="https://agentmods.dev/skills/evoelsewhere/evoflux/super-research"><img src="https://agentmods.dev/badge/skills/evoelsewhere/evoflux/super-research/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/evoelsewhere/evoflux/super-research"><img src="https://agentmods.dev/badge/skills/evoelsewhere/evoflux/super-research.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 | $0.00199 | $0.02671 |
| Opus 5.5 | $0.00080 | $0.01068 |
| Sonnet 5.5 | $0.00040 | $0.00534 |
| Haiku 4.5 | $0.00020 | $0.00267 |
Grade A, and why
super-research 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 1mo 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.
This is a copy
89% identical to super-research — 94 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Autonomous Research
The value of this skill is not the specific procedure — it is the property that research work done under it is comparable, honest, and auditable. Ten cheap experiments, queries, or analyses done to the same standard beat one clever untested claim. This is what lets a human check in eight hours later and trust what they see.
The method is distilled from Karpathy's autoresearch and generalized to eight research modes.
Shared discipline (all modes)
Every mode operates under the same five rules. Read them before branching into a mode.
-
State a contract before you begin. Infer everything you can from the workspace and the request, then write down the goal, the primary output, and the stopping condition, and get one confirmation. This is your last question. After confirmation you are autonomous.
-
Establish a baseline as your first artifact. Every mode has a version of "the answer without any of my work" — the unmodified code, the first three sources found, the raw dataset before any transformation. Record it first. Without a baseline, "better" and "significant" are meaningless.
-
Log every step in a machine-readable file, including failures. A tab-separated log (TSV, not CSV — descriptions contain commas) with a header row and one row per attempt. Failed attempts get a
crash/dead-end/inconclusivestatus. Silently discarding attempts is the fastest way to fool yourself and the user; the log is the evidence that the work happened. -
Never pause to ask permission mid-loop. Once the contract is confirmed, do not stop to check in, propose stopping at a "natural break", or ask "should I keep going?". The human may be asleep and expects to wake up to a full log. The loop ends only at the agreed stopping condition or a manual interruption. This is the single most common failure mode of autonomous runs.
-
Never game the metric, the sources, or the analysis. Do not edit the eval code because it is "clearly wrong". Do not drop a source because it does not fit the thesis. Do not p-hack or cherry-pick a subset that gives a nicer number. If the ground truth looks wrong, log the concern in the description column and keep going — the human decides later.
What ships with it
21 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/ablation-study.md 8.1 KB
- references/api-cheatsheet.md 3.0 KB
- references/benchmark-comparison.md 5.5 KB
- references/experiment-loop.md 5.2 KB
- references/paper-reproduction.md 7.8 KB
- references/paper-writing.md 9.6 KB
- references/paper-writing/abstract.md 3.1 KB
- references/paper-writing/conclusion.md 1.2 KB
- references/paper-writing/experiments.md 4.6 KB
- references/paper-writing/introduction.md 13 KB
- references/paper-writing/method.md 6.2 KB
- references/paper-writing/paper-review.md 5.3 KB
- references/paper-writing/paragraph-flow.md 2.1 KB
- references/paper-writing/pdf-export.md 4.0 KB
- references/paper-writing/related-work.md 1.3 KB
- references/quant-analysis.md 6.9 KB
- references/root-cause.md 8.2 KB
- references/topic-survey.md 9.9 KB
- scripts/fetch_paper.py 7.3 KB runs code
- scripts/paper_search.py 7.3 KB runs code
- scripts/verify_citation.py 9.0 KB runs code
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.
- 1mo ago Changed · +16 lines · +84 tokens per session a9273beb73ac
- 1mo ago Changed · +13 tokens per session 4e807e714deb
- 1mo ago First seen · 91 lines · 102 tokens per session scan A 799b940477e6
super-research is a skill published in the GitHub repository evoelsewhere/evoflux (13 stars, last pushed yesterday), licensed Apache-2.0. It adds 199 tokens to every session and 2,671 once invoked, about $0.0008 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 89% identical to super-research, differing in 94 lines, and is treated as a copy.
Other skills, from other repositories
academic-paper-expert
An academic writing guide for planning, drafting, reviewing, and polishing research papers. It covers common paper sections, literature reviews, abstracts, and APA or GB/T citation formats.
science-writing-expert
A science-communication writing guide that explains scientific ideas, technology reviews, and research news in everyday language. It uses analogies and step-by-step explanations while marking uncertainty and correcting common misunderstandings.
cuopt-numerical-optimization-formulation
LP, MILP, QP — concepts, problem-text parsing, and formulation patterns (parameters, constraints, decisions, objective). Concepts only; no API.
nemo-mbridge-perf-expert-parallel-overlap
Validate and use MoE expert-parallel communication overlap in Megatron-Bridge, including overlapmoeexpertparallelcomm, delaywgradcompute, and flex dispatcher backends such as DeepEP and HybridEP.
msa-structure-prediction-pipeline
NOTE: your protein sequence and the retrieved MSA alignment are transmitted to external NVIDIA-hosted APIs (health.api.nvidia.com) on every call. Use local NIM containers for confidential or proprietary sequences. Run a complete protein structure prediction pipeline using NVIDIA BioNeMo NIMs: search for MSA alignments…
demo-3-repo-pulse
Demo skill that pulls live GitHub API data for a repo, analyses a full year of weekly commit activity and top contributors, and renders a single-file dashboard.html with stat tiles and inline SVG charts — no chart library.