AgentSkillsCN

legacy-survey-api

通过查看器URL API与发布文件结构,结合本地缓存刷新流程、maskbits解读,以及已知问题的排查,为Legacy Survey DR10数据(图像与目录)的使用提供指导。在涉及裁剪、拖拉机/扫掠目录、砖块,或为Legacy Survey数据产品进行质量掩码处理的任务中使用。

SKILL.md
--- frontmatter
name: legacy-survey-api
description: "Guidance for using the Legacy Survey DR10 data (images and catalogs) through viewer URL APIs and release file structure, with local cache refresh workflows, maskbits interpretation, and known-issues checks. Use when tasks involve cutouts, tractor/sweep catalogs, bricks, or quality masking for Legacy Survey data products."

Legacy Survey API

Use this skill for DR10 image/catalog access and interpretation while keeping context usage small.

Workflow

  1. Confirm the target data product: image cutout, brick-level tractor catalog, or sweep catalog query.
  2. Refresh local references before analysis:
    • python3 scripts/refresh_legacysurvey_cache.py --output-dir references/cache
    • For no-network environments, use --dry-run and rely on bundled references.
  3. Load only the minimum reference files from references/index.md needed for the task.
  4. Build API URLs using references/viewer_api_urls.md patterns.
  5. Validate mask flags and caveats using references/dr10_bitmasks.md and references/dr10_issues.md before final conclusions.

Keep Context Lean

  • Do not load all reference files at once.
  • Start from references/index.md, then open only relevant files.
  • Prefer catalog-specific references for catalog tasks and image-specific references for cutout tasks.

References

  • references/index.md
  • references/dr10_status.md
  • references/dr10_description.md
  • references/dr10_files.md
  • references/dr10_catalogs.md
  • references/dr10_bitmasks.md
  • references/dr10_issues.md
  • references/viewer_api_urls.md