Automate Flatbrowser dataset upkeep with qualification, enrichment, and area-research rules.
npx skills add https://github.com/Capataina/Flat-Browser --skill data-upkeep
Please help me install this Agent Skill. Skill: data-upkeep Source: https://github.com/Capataina/Flat-Browser/tree/main/.claude/skills/data-upkeep Command: npx skills add https://github.com/Capataina/Flat-Browser --skill data-upkeep
Flatbrowser's dataset requires regular validation and enrichment to prevent stale or incorrect entries from degrading decision quality. This Skill orchestrates end-to-end upkeep across operators and areas, applying qualification, enrichment, area research, and calibration workflows to maintain integrity and support reliable user decisions.
Invoke the data-upkeep skill to refresh operator and area data, generate batch manifests, and prepare proposals for review.
High-intent search queries and answers about installing and using this skill.
Dataset upkeep automation applies qualification, enrichment, and area-research rules to prevent stale entries. It orchestrates end-to-end validation across operators and areas within a controlled workflow to maintain dataset integrity.
Ghost-project detection identifies invalid or duplicate project entries within a dataset. It flags these ghost projects for reattribution or removal, using explicit user confirmation gates to ensure accurate data calibration before changes are committed.
Relative calibration across data batches is maintained using batch manifests, comparables injection, and cross-batch review. This phased gating approach ensures consistent dataset accuracy when processing separate operator and area batches.
Yes, research proposals with explicit sources are generated by writing structured context files during the enrichment workflow. The process enforces validation and optional recalibration to ensure proposed dataset updates are fully sourced and gated for confirmation.
Yes, the workflow operates on both operator and area batches simultaneously. It applies area-research rules and cross-batch review to keep 266 projects and 55 areas accurate without requiring separate manual processing cycles.
Automated data enrichment requires explicit user confirmation for ghost-detection, reattribution, and missing-project additions. This gating mechanism prevents unintended modifications, meaning large datasets cannot be updated without manual review of flagged issues.
seanpjones-collabFilter nationwide buyer data by county and state into an Excel workbook with decision-maker fields.
inbrainfunAutomate a 7-phase pipeline to extract, deduplicate, and update canonical tracker datasets.
nimeshgurungExecute implementation plans in batches with review checkpoints between each.
codataConvert raw data sources into Croissant JSON-LD dataset metadata.
