Index ADVPL/TLPP source files into a local SQLite database with FTS5.
npx skills add https://github.com/JoniPraia/plugadvpl --skill plugadvpl-ingest
Please help me install this Agent Skill. Skill: /plugadvpl:ingest Source: https://github.com/JoniPraia/plugadvpl/tree/main/skills/ingest Command: npx skills add https://github.com/JoniPraia/plugadvpl --skill plugadvpl-ingest
This skill eliminates slow, token-heavy reading of large ADVPL/TLPP codebases by indexing source structure into a local SQLite database you can query efficiently.
.plugadvpl/index.db for fast lookups of functions, tables, and related metadata.Run ingest on your repository by telling the AI to execute: uvx [email protected] ingest ..
High-intent search queries and answers about installing and using this skill.
The incremental ingest feature reindexes only changed ADVPL and TLPP files and issues correctness warnings if lookup rules changed, ensuring safe partial directory indexing without skipping unsafe updates.
You can redact secrets during ADVPL codebase ingestion by passing the `--redact-secrets` flag, which ensures sensitive information is safely handled before writing source metadata into the local SQLite database.
Yes, you can use parallel workers to speed up ADVPL repository ingestion by specifying the `--workers` option, which enables parallel processing of source files during the indexing operation.
Indexing ADVPL and TLPP sources into a local SQLite database eliminates slow, token-heavy reading by allowing you to query the indexed structure directly for fast lookups of functions and tables.
Yes, the ADVPL indexing tool supports incremental reindexing through the `--incremental` flag, with correctness warnings that detect when lookup rules changed and skipping became unsafe.
lis186Search a repository for code patterns and return matching files with line numbers.
bahayonghangIndex Markdown notes into a local SQLite vector/full-text index.
ItsCodejacManage AI coding session data by reclaiming disk space, securing secrets, and recovering lost work.
jflambBuild local SQLite knowledge stores with FTS5 full-text search.