aws-sdk-pandas · VectorSpaceLab - Agent Skill

01 前言 Front Matter

name: aws-sdk-pandas
license: Apache 2.0
metadata: { "disco-role": "operating" }
description: Routes AWS SDK for pandas (awswrangler) workflows across S3 lakehouse data, Athena/Glue catalogs, SQL connectors, runtime configuration, and AWS service integrations.
disable-model-invocation: true

02 资源文件 5

参考文档references/repo-provenance.md1.3 KB
参考文档references/repo-routing-metadata.json407 B
参考文档references/runtime-overview.md3.7 KB
参考文档references/troubleshooting.md3.3 KB
脚本scripts/check_runtime.py5.8 KB

03 SKILL.md 正文

AWS SDK for pandas

Use this skill when a request names awswrangler, aws-sdk-pandas, Pandas on AWS, or a wr.* API from the repository.

This repo is a Python API package, not a CLI tool. The first questions are usually:

how to install the right extras,
how to import and verify the runtime,
which AWS service family the user is working with,
and which AWS credentials, region, or optional backend are required.

Read references/runtime-overview.md first for install, extras, and runtime facts. Read references/troubleshooting.md when import errors, missing extras, region issues, or AWS service prerequisites are blocking you. Read references/repo-provenance.md before deciding whether this skill is current for the checkout. Run scripts/check_runtime.py after installation to confirm importability and optional dependency availability.

Route map

If the user asks about...	Read this sub-skill
Installing, importing, setting `WR_*` env vars, switching `wr.engine` / `wr.memory_format`, or debugging missing extras	`sub-skills/runtime-and-config/SKILL.md`
`wr.s3.*` object, file-format, dataset, S3 Tables, or S3 Vectors workflows	`sub-skills/s3-lakehouse/SKILL.md`
`wr.catalog.`, `wr.athena.`, `wr.data_quality.`, or `wr.cleanrooms.` workflows	`sub-skills/catalog-and-query/SKILL.md`
`wr.redshift.`, `wr.data_api.`, `wr.mysql.`, `wr.postgresql.`, `wr.sqlserver.`, or `wr.oracle.` workflows	`sub-skills/sql-connectors/SKILL.md`
`wr.dynamodb.`, `wr.timestream.`, `wr.opensearch.`, `wr.neptune.`, `wr.quicksight.`, `wr.cloudwatch.`, `wr.emr.`, `wr.emr_serverless.`, `wr.sts.`, `wr.secretsmanager.`, or `wr.chime.*` workflows	`sub-skills/service-integrations/SKILL.md`

Shared operating pattern

Confirm the runtime with scripts/check_runtime.py when the package was just installed or the environment changed.
Pick the service-family sub-skill that matches the wr.* API surface.
Use wr.config and boto3.Session for defaults, region, and credentials.
Install only the extras needed for the chosen workflow family.
Keep live AWS operations separate from local smoke checks; if a workflow needs credentials or managed AWS resources, say so early.

Common signals

NoCredentialsError or NoRegionError usually means the AWS session is not configured.
Missing optional dependency ... usually means the relevant extra was not installed.
Ray/Modin questions belong in runtime-and-config even when they affect wr.s3 or wr.athena calls.
Live AWS service prerequisites and IAM needs belong in the nearest service sub-skill, not here.

Shared references and scripts

references/runtime-overview.md for install commands, supported versions, and optional extras.
references/troubleshooting.md for cross-cutting import, dependency, and credential failures.
scripts/check_runtime.py for a safe import/config/extra check.

04 同仓库更多 Skills

VectorSpaceLab/AREX-Skill

passnet-triton-opt

Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block/grid/warp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.

★ 351

VectorSpaceLab/AREX-Skill

passnet-solve

END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.

★ 351

VectorSpaceLab/AREX-Skill

passnet-skill

PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under ./pass_dir/, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.

★ 351

VectorSpaceLab/AREX-Skill

passnet-pattern-fusion

Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()/replacement_args()/replacement_func() files, when a pass "failed to match", or when deciding how to fuse multiple kernels/ops.

★ 351

VectorSpaceLab/AREX-Skill

passnet-orchestrate

Round/batch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep/revert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.

★ 351

VectorSpaceLab/AREX-Skill

passnet-feedback

Fast, deterministic feedback for PassNet work: pre-flight pattern/match verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).

★ 351