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 theam/limina --skill exploratory-sota-researchgit clone --depth 1 https://github.com/theam/liminaWrote 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/theam/limina/exploratory-sota-research)<a href="https://agentmods.dev/skills/theam/limina/exploratory-sota-research"><img src="https://agentmods.dev/badge/skills/theam/limina/exploratory-sota-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/theam/limina/exploratory-sota-research"><img src="https://agentmods.dev/badge/skills/theam/limina/exploratory-sota-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 | $0.00048 | $0.01744 |
| Opus 5 | $0.00024 | $0.00872 |
| Sonnet 5 | $0.00010 | $0.00349 |
| Haiku 4.5 | $0.00005 | $0.00174 |
Grade A, and why
exploratory-sota-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 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 — 278 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Exploratory SOTA Research
Use this skill to map the external mechanism landscape for a hard AI/ML research problem and turn that map into concrete Limina artifacts.
When to use it
Use this skill when the user wants any of the following:
- an exploratory literature search
- a state-of-the-art map
- a mechanism landscape for a technical problem
- candidate research directions for AI/ML/DL/LLM systems
- promising approaches to improve a bottleneck
- a benchmark, reproducibility, or artifact scan
- a recommendation about which technical directions are worth testing
Do not use this skill for:
- simple factual lookup
- single-paper summarization when no broader research mapping is needed
- generic brainstorming with no research component
- implementation-only requests
- questions where the user clearly wants a direct answer rather than an exploratory research process
Limina adapter
When you use this skill inside a Limina repo:
- Read
kb/mission/CHALLENGE.md. - Read
kb/ACTIVE.md. - Open only the linked artifacts you need next.
- Persist serious external sources as
Lnotes inkb/research/literature/. - If the landscape changes the direction, update or create the relevant
H,CR,SR, andACTIVE.mdentries instead of leaving the conclusion only in chat. - Prefer
python3 scripts/kb_new_artifact.py ...when a new core artifact is needed.
The goal is not just to produce a good survey. The goal is to change the research graph in a grounded way.
Non-negotiable principles
-
Solve the real decision problem. Infer the objective, constraints, and decision context behind the request.
-
Search by challenge and mechanism. Convert the task into problem slices, failure modes, root causes, and mechanism families.
-
Prefer generalizable solutions. Prioritize mechanisms that transfer across datasets, tasks, or domains unless the user explicitly wants a narrow patch.
-
Evidence beats hype. Reward realistic baselines, meaningful gains, strong ablations, transparent limitations, reproducibility signals, and public artifacts.
What ships with it
5 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.
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 · 278 lines · 48 tokens per session scan A 0fd490f43b48
exploratory-sota-research is a skill published in the GitHub repository theam/limina (40 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 48 tokens to every session and 1,744 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
dspy-gepa-reflective
Use for GEPA reflective optimization, ReAct agent optimization, feedback metrics, LLM reflection, and execution trajectories.
human-agent-trust-exploit-detection
Detect social engineering, deceptive responses, false assurances, or prompts that induce unsafe user actions.
shap
Model interpretability and explainability using SHAP (SHapley Additive exPlanations). Use this skill when explaining machine learning model predictions, computing feature importance, generating SHAP plots (waterfall, beeswarm, bar, scatter, force, heatmap), debugging models, analyzing model bias or fairness, comparing…
glycobiology
Glycosylation site prediction and glycobiology analysis. N-glycosylation motif finding, O-glycosylation hotspot prediction, glycan structure resources. Lightweight, pure Python. For protein function queries use uniprot-database; for structure analysis use alphafold-database.
cellxgene-census
Query the CELLxGENE Census (61M+ cells) programmatically. Use when you need expression data across tissues, diseases, or cell types from the largest curated single-cell atlas. Best for population-scale queries, reference atlas comparisons. For analyzing your own data use scanpy or scvi-tools.
esm
Comprehensive toolkit for protein language models including ESM3 (generative multimodal protein design across sequence, structure, and function) and ESM C (efficient protein embeddings and representations). Use this skill when working with protein sequences, structures, or function prediction; designing novel…