Borrowing it
Nothing to install: this file belongs to tensorcircuit/tensorcircuit-ng. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/tensorcircuit/tensorcircuit-ng/master/.agents/skills/meta-explorer/SKILL.mdgit clone --depth 1 https://github.com/tensorcircuit/tensorcircuit-ngWrote 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/tensorcircuit/tensorcircuit-ng/meta-explorer)<a href="https://agentmods.dev/skills/tensorcircuit/tensorcircuit-ng/meta-explorer"><img src="https://agentmods.dev/badge/skills/tensorcircuit/tensorcircuit-ng/meta-explorer/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/tensorcircuit/tensorcircuit-ng/meta-explorer"><img src="https://agentmods.dev/badge/skills/tensorcircuit/tensorcircuit-ng/meta-explorer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00060 | $0.01221 |
| Opus 5.5 | $0.00024 | $0.00488 |
| Sonnet 5.5 | $0.00012 | $0.00244 |
| Haiku 4.5 | $0.00006 | $0.00122 |
Grade A, and why
meta-explorer 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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
When acting as a Meta-Explorer, you are an autonomous researcher tasked with discovering the optimal quantum circuit or optimization strategy for a given objective. This objective can span multiple domains:
- VQE: Ground state energy minimization for physics models.
- QAOA: Combinatorial optimization on graphs.
- QML: Classification, regression, or generative modeling using Variational Quantum Classifiers (VQC).
- Non-VQA: Quantum state tomography, circuit compression and compilation, or error mitigation strategy discovery, etc.
1. Workspace Initialization
- Create Directories: Initialize the research folder:
examples/meta_exploration/<YYYYMMDD>_<short_objective>/. - Create Subfolders: Create
.snapshots/to store code for every single successful experiment. - Define
objective.py: Create a fixed script that contains:- The problem core (Hamiltonian, Dataset, or Target State).
- A
evaluate(circuit_fn, params)method that returns the core metric (e.g., energy, fidelity, loss, or accuracy).
- Initialize
ledger.json: Create a file to track metadata, results, and "Agent Thoughts" for every experiment.
2. Multi-Frontier Exploration Loop
You must maintain a Top-K Frontier (default K=3) of the most promising but diverse approaches.
Exploration Intensity: Push very hard to explore at least 30+ different experiments (counting variants in hyperparameters, topologies, and initializations) before declaring a winner. If you can still observe evident progress after these experiments, don't stop, continue to push and explore.
Creative Search:
- Beyond Depth & Optimizer: Do NOT limit your search to simply increasing layers or swapping optimizers. Rethink the problem's fundamental structure.
- Internal Brilliance: Be creative! Invent new gate patterns, explore non-native lattice connectivities, or use ancilla-assisted measurement schemes, try su(4) two qubit gates which are the most expressive.
- Literature Review: Use search tools to find promising ideas in recent quantum computing literature and port them to TensorCircuit.
- Reference Examples: Examine existing scripts in the repository for implementation patterns and best practices.
What ships with it
1 file 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.
- 2mo ago First seen · 65 lines · 60 tokens per session scan A f9dfd17dee54
meta-explorer is a skill published in the GitHub repository tensorcircuit/tensorcircuit-ng (93 stars, last pushed today), licensed Apache-2.0. It adds 60 tokens to every session and 1,221 once invoked, about $0.0002 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.
Other skills, from other repositories
physicsnemo-discover
Official NVIDIA-authored guidance for navigating PhysicsNeMo — pick the model, datapipe, or example for a SciML/AI4Science task (surrogates, forecasting, downscaling, physics-informed, inverse, generative). Points at existing files via live repo search; never writes code. Do NOT use for installation or environment…
cellxgene-census-query
Query CZ CELLxGENE Census (61M+ cells). Filter by cell type/tissue/disease, retrieve expression data, and integrate with scanpy/PyTorch for population-scale single-cell analysis. Use this skill when: (1) Querying single-cell expression data by cell type, tissue, or disease, (2) Exploring available single-cell datasets…
physicsnemo-shard-tensor
Official NVIDIA-authored guidance for PhysicsNeMo ShardTensor domain parallelism — integrate domain parallelism into training/inference scripts (new or existing) with DDP or FSDP2, write and register shard patches to enable new layers/ops, and bootstrap multi-GPU correctness tests. Use when working with ShardTensor…
fine-tuning-serving-openpi
Fine-tune and serve Physical Intelligence OpenPI models (pi0, pi0-fast, pi0.5) using JAX or PyTorch backends for robot policy inference across ALOHA, DROID, and LIBERO environments. Use when adapting pi0 models to custom datasets, converting JAX checkpoints to PyTorch, running policy inference servers, or debugging…
ml-training-recipes
Battle-tested PyTorch training recipes for all domains — LLMs, vision, diffusion, medical imaging, protein/drug discovery, spatial omics, genomics. Covers training loops, optimizer selection (AdamW, Muon), LR scheduling, mixed precision, debugging, and systematic experimentation. Use when training or fine-tuning…
pysr
Use when fitting equations to data with PySR or SymbolicRegression.jl, when a user wants an interpretable formula, symbolic model, scaling law, or empirical relation discovered from numeric data, or when debugging a PySR search that is slow, stuck, or giving poor equations.