pufferlib skill

by K-Dense-AI·K-Dense-AI/scientific-agent-skills·47k stars·MIT

Version-aware guidance for PufferLib reinforcement-learning environments, vectorization, policies, PuffeRL training, evaluation, and safe checkpoint review. Use when adapting Gymnasium/PettingZoo environments to published PufferLib 3.0.0 or working with the redesigned native 4.0 source line.

A100/100content scan

Is the pufferlib skill safe?

Clean: nothing in its files matched our rules. We read 15 files in the folder on 2026-09-28.

No findings.

Install the pufferlib skill

A skill is a folder. Copy it into your agent's skills folder and the agent loads it when the task matches its description.

git clone --depth 1 https://github.com/K-Dense-AI/scientific-agent-skills.git /tmp/scientific-agent-skills
mkdir -p ~/.claude/skills
cp -r /tmp/scientific-agent-skills/skills/pufferlib ~/.claude/skills/pufferlib

available in every project

In the Claude apps, zip the folder and upload it from the Skills settings. The folder on GitHub

The instructions your agent would load

SKILL.md as published, without the frontmatter. Read it on GitHub

PufferLib

Use PufferLib with an explicit version profile. Upstream currently has two incompatible surfaces:

Do not combine 3.0 imports with 4.0 config/CLI examples. The 4.0 redesign removed the 3.0 emulation, vector, and pytorch modules from the current package tree.

Safe defaults

only allowlisted built-ins and slug identifiers.

Start with bundled synthetic, CPU-only, network-free tools.
Do not import an arbitrary environment by dotted path. Bundled tools accept

extension, ROM, map, checkpoint, or pickle file.

Do not install or execute an unreviewed environment package, native

attestations, and build hooks. Sandbox native builds and first execution.

Verify official source, immutable revision, licenses, checksums or

render size, and wall time.

Cap steps, environments, agents, workers, threads, buffers, memory, disk,

disclosure acknowledgment, and separate artifact-upload approval.

Keep training and evaluation environments/seeds separate.
Default logging to local/none. External logging requires explicit opt-in,

logger configuration. Never print them.

Never pass W&B or Neptune credentials via CLI, INI, JSON, tags, run names, or

is not proof of safety.

Never dump all environment variables or recursively search for .env.
Hash checkpoint bytes before trusted, sandboxed loading; metadata inspection

First local checks

All bundled CLIs are dependency-free and emit strict JSON:

python3 scripts/env_template.py --help
python3 scripts/env_contract_validator.py
python3 scripts/benchmark_vectorization.py --backend serial
python3 scripts/train_template.py
python3 scripts/validate_plan.py
python3 scripts/repro_plan.py

Defaults are synthetic, deterministic, bounded, local, CPU-only, no-network, and dry-run where training would otherwise occur.

Installation and provenance

Published 3.0.0

PyPI supplies only pufferlib-3.0.0.tar.gz:

sha256: 7df3a3e3f5f894d78d2a1f5374097890aec01473183e748abefe4f3faa10eaa9
Requires-Python: >=3.9

After source/build review, create a pinned uv project:

uv venv --python 3.11
uv add --exact --no-sync "pufferlib==3.0.0"
uv lock
uv sync --frozen

Commit pyproject.toml and uv.lock; verify the archive digest and every resolved dependency. The source build can compile native code and fetch build assets, so resolve/build in a sandbox without credentials or sensitive mounts. The uploaded metadata does not pin Torch or CUDA; do not claim a supported CUDA matrix that PyPI does not declare.

Current 4.0 source

The reviewed branch head on 2026-07-23 was:

25647630e1b15330bb3153a5a0d3ff8d234c3acf

Pin the commit, not branch 4.0:

uv add --no-sync \
  "pufferlib @ git+https://github.com/PufferAI/PufferLib.git@25647630e1b15330bb3153a5a0d3ff8d234c3acf"
uv lock

The current package declares Python >=3.10 and Torch >=2.9. Upstream PufferTank currently uses Ubuntu 24.04, Python 3.12, and an NVIDIA CUDA 13.0.2/cuDNN development image with the cu130 Torch index, but does not pin the exact Torch wheel or all system packages. Treat it as a reference, not a complete lock. Never execute a remote installer directly from a pipe.

Read references/training.md before any installation or build.

Environment workflow

1. Validate the contract

Gymnasium reset returns (observation, info). Step returns:

(observation, reward, terminated, truncated, info)

Validate spaces, shapes, dtypes, finite rewards, booleans, reset-before-step, reset-after-end, seeding, and cleanup. terminated is an MDP terminal; truncated is an external cutoff such as a time limit. Preserve the distinction for bootstrapping and metrics.

python3 scripts/env_contract_validator.py \
  --steps 64 --episodes 8 --seed 42

2. Adapt only after review

Published 3.0 uses explicit wrappers:

import pufferlib.emulation

wrapped = pufferlib.emulation.GymnasiumPufferEnv(reviewed_gymnasium_instance)

For a reviewed PettingZoo Parallel environment:

wrapped = pufferlib.emulation.PettingZooPufferEnv(reviewed_parallel_instance)

There is no supported 3.0 pufferlib.emulate(...) shortcut matching the old skill. Read references/environments.md and references/integration.md.

3. Native environments

Published 3.0 PufferEnv requires singleobservationspace, singleactionspace, and numagents before super().init__(buf). It uses in-place vector buffers and returns separate terminal/truncation arrays plus a list of info dictionaries.

Current 4.0 uses C bindings. Start from upstream ocean/squared (single-agent) or ocean/target (multi-agent), build one environment in local/sanitized mode, and verify every buffer size/type/index before optimization.

Vectorization workflow

Published 3.0:

import pufferlib.vector

vecenv = pufferlib.vector.make(
    reviewed_creator,
    backend=pufferlib.vector.Serial,
    num_envs=4,
    seed=42,
)

Move to Multiprocessing only after serial traces pass. Record numenvs, numworkers, batchsize, zero-copy mode, start method, agent count, masks, and actual returned shapes. For multi-agent environments, batch length is based on agent slots, not necessarily numenvs.

Current 4.0 config instead uses:

[vec]
total_agents = 4096
num_buffers = 2
num_threads = 16

Read references/vectorization.md. Benchmark fixed work with warmup and at least three repeats; report simulation and end-to-end training SPS separately. The bundled benchmark measures only its synthetic harness.

Policy workflow

Published 3.0 policies are Torch modules sized from singleobservationspace/singleactionspace. Stable recurrent composition uses encodeobservations and decodeactions; structured emulation uses pufferlib.pytorch.nativizedtype and nativizetensor.

More skills from K-Dense-AI/scientific-agent-skills

AadaptyvHow to use the Adaptyv Bio Foundry API and Python SDK for protein experiment design, submission, and results retrieval. Use this skill whenever the user mentions Adaptyv, Foundry API, protein binding assays, protein screening experiments, BLI/SPR assays, thermostability assays, or wants to submit protein sequences for experimental characterization. Also trigger when code imports `adaptyv`, `adaptyv_sdk`, or `FoundryClient`, or references `foundry-api-public.adaptyvbio.com`.
AaeonThis skill should be used for time series machine learning tasks including classification, regression, clustering, forecasting, anomaly detection, segmentation, and similarity search. Use when working with temporal data, sequential patterns, or time-indexed observations requiring specialized algorithms beyond standard ML approaches. Particularly suited for univariate and multivariate time series analysis with scikit-learn compatible APIs.
AalphagenomeLook up precomputed AlphaGenome Atlas effects for any GRCh38 single-nucleotide variant (AVI score with Phred and 18 SHAP feature attributions, plus raw and quantile scores for RNA-seq, DNase, ATAC, ChIP-TF, ChIP-histone, CAGE, PRO-cap, splicing, polyadenylation and contact-map tracks), score variants or scan windows on demand with the AlphaGenome model for human and mouse (variant scoring, in silico mutagenesis, REF-versus-ALT track prediction), and build Atlas website deep links. Use when the user mentions AlphaGenome, AlphaGenome Atlas, AVI or AlphaGenome Variant Impact, DeepMind variant effect prediction, or wants to prioritise or mechanistically interpret non-coding, regulatory, splicing, enhancer, promoter, or chromatin-accessibility effects of SNVs from a VCF, credible set, or region. Research use only; not a clinical tool.
Aanalytical-method-validationPlan, execute, and document validation, verification, and transfer of analytical procedures under the governing framework - ICH Q2(R2) and Q14, USP <1220>/<1225>/<1226>, ICH M10 bioanalytical, CLSI EP, or ISO/IEC 17025. Use for HPLC, LC-MS/MS, GC, CE, ICP-MS, dissolution, qNMR, qPCR, NIR, and ligand binding or cell-based assays whenever the question is whether a procedure is fit for its intended purpose. Triggers include "method validation", "analytical method validation", "AMV", "validation protocol", "acceptance criteria", "linearity", "reportable range", "accuracy and precision", "repeatability", "intermediate precision", "recovery", "LOD", "LOQ", "detection limit", "quantitation limit", "specificity", "robustness", "method transfer", "method comparison", "Deming", "Passing-Bablok", "Bland-Altman", "equivalence testing", "OOS investigation", "ICH Q2", "Q2(R2)", "Q14", "USP 1225", "ICH M10", "incurred sample reanalysis", "ISR", "CLSI EP", and any request to show that an assay works.
AanndataData structure for annotated matrices in single-cell analysis. Use when working with .h5ad files or integrating with the scverse ecosystem. This is the data format skill—for analysis workflows use scanpy; for probabilistic models use scvi-tools; for population-scale queries use cellxgene-census.
AarborAutonomously improve a real artifact (code, training recipe, agent harness, data pipeline, prompt) against an objective and an evaluator, using Hypothesis Tree Refinement (HTR) from the Arbor paper. Use this whenever someone wants to iteratively optimize something over many experiments without overfitting — e.g. "get my model's eval score up", "improve this agent/harness", "tune this pipeline", "beat the baseline on this benchmark", "run a search over approaches and keep the best", "do an MLE-bench / Kaggle-style optimization", or any long-horizon "make this artifact better and don't just memorize the dev set" task. Trigger it even when the user doesn't say "Arbor" or "hypothesis tree" but describes repeated experiment-and-evaluate loops, branching exploration of competing ideas, or worries about a dev/test gap. Runs Claude itself as the coordinator with subagent executors in isolated git worktrees; for the standalone `arbor` CLI tool see references/arbor-upstream.md.
AarboretoInfer gene regulatory networks (GRNs) from gene expression data using scalable algorithms (GRNBoost2, GENIE3). Use when analyzing transcriptomics data (bulk RNA-seq, single-cell RNA-seq) to identify transcription factor-target gene relationships and regulatory interactions. Supports distributed computation for large-scale datasets.
AastropyCore Python library for astronomy and astrophysics workflows that need Astropy APIs, including units/quantities, coordinates, FITS I/O, tables, time systems, WCS, and cosmology. Use when implementing or debugging astronomical data analysis code with Astropy.
AautoskillObserve the user's screen via screenpipe, detect repeated research workflows, match them against existing scientific-agent-skills, and draft new skills (or composition recipes that chain existing ones) for the patterns not yet covered. Use when the user asks to analyze their recent work and propose skills based on what they actually do. Requires the screenpipe daemon (https://github.com/screenpipe/screenpipe) running locally on port 3030 — the skill has no other data source and will refuse to run if screenpipe is unreachable. All detection runs locally; only redacted cluster summaries reach the LLM.
Abenchling-integrationBenchling Python SDK and REST API integration for registry entities, inventory, ELN entries, workflows, Benchling Apps, and Data Warehouse queries. Use when automating lab data with benchling-sdk or the v2 API.
Abgpt-paper-searchSearch scientific papers and retrieve structured experimental data extracted from full-text studies via the BGPT MCP server. Returns 25+ fields per paper including methods, results, sample sizes, quality scores, and conclusions. Use for literature reviews, evidence synthesis, and finding experimental details not available in abstracts alone.
AbidsUse this skill when working with Brain Imaging Data Structure (BIDS) datasets: organizing neuroscience and biomedical data (MRI, EEG, MEG, iEEG, PET, microscopy, NIRS, motion capture, EMG, MR spectroscopy, behavioral), querying BIDS layouts, validating compliance, converting DICOM to BIDS, writing metadata sidecars, or creating BIDS derivatives.

All agent skills → · MCP servers