Memory

RAGFlow

Manage RAGFlow datasets, documents, parsing, chunk retrieval, and configured model listings.

What it does

Manage RAGFlow datasets and their documents through bundled scripts, including creation, inspection, updates, uploads, deletion, and model listing. Start or stop document parsing, report parse status and API error fields exactly, and retrieve chunks across selected datasets or documents. Destructive actions require resolved IDs, an exact item list, and explicit confirmation.

When to use it

  • Managing RAGFlow datasets
  • Uploading and maintaining dataset documents
  • Monitoring document parsing
  • Retrieving chunks from selected datasets

The skill document

RAGFlow Dataset And Retrieval

Use only the bundled scripts in scripts/. Prefer --json so returned fields can be relayed exactly. Follow reference.md for all user-facing output.

Use This Skill When

  • the user wants to create, list, inspect, update, or delete RAGFlow datasets
  • the user wants to upload, list, update, or delete documents in a dataset
  • the user wants to start parsing, stop parsing, or check parse progress
  • the user wants to retrieve chunks from one or more datasets
  • the user wants to list configured RAGFlow models

Core Workflow

  1. Resolve the target dataset or document IDs first.
  2. Run the matching script from scripts/.
  3. Use --json unless a script only needs a simple text response.
  4. Return API fields exactly; do not guess missing details.

Common commands:

python3 scripts/datasets.py list --json
python3 scripts/datasets.py info DATASET_ID --json
python3 scripts/datasets.py create "Example Dataset" --description "Quarterly reports" --json
python3 scripts/update_dataset.py DATASET_ID --name "Updated Dataset" --json
python3 scripts/upload.py DATASET_ID /path/to/file.pdf --json
python3 scripts/upload.py list DATASET_ID --json
python3 scripts/update_document.py DATASET_ID DOC_ID --name "Updated Document" --json
python3 scripts/parse.py DATASET_ID DOC_ID1 [DOC_ID2 ...] --json
python3 scripts/stop_parse_documents.py DATASET_ID DOC_ID1 [DOC_ID2 ...] --json
python3 scripts/parse_status.py DATASET_ID --json
python3 scripts/search.py "query" --json
python3 scripts/search.py "query" DATASET_ID --json
python3 scripts/search.py --dataset-ids DATASET_ID1,DATASET_ID2 --doc-ids DOC_ID1,DOC_ID2 "query" --json
python3 scripts/search.py --retrieval-test --kb-id DATASET_ID "query" --json
python3 scripts/list_models.py --json

Guardrails

  • For any delete action, list the exact items first and require explicit user confirmation before executing.
  • Delete only by explicit dataset IDs or document IDs. If the user gives names or fuzzy descriptions, resolve IDs first.
  • Upload does not start parsing. Start parsing only when the user asks for it.
  • parse.py returns immediately after the start request; use parse_status.py for progress.
  • For progress requests, use parse_status.py on the most specific scope available:
    • dataset specified: inspect that dataset
    • document IDs specified: pass --doc-ids
    • no dataset specified: list datasets first, then aggregate status across datasets
  • If a parse status result includes progress_msg, surface it directly. For FAIL, treat it as the primary error detail.
  • Use --retrieval-test only for single-dataset debugging or when the user explicitly asks for that endpoint.

Output Rules

  • Follow reference.md.
  • Use tables for 3+ items when possible.
  • Preserve api_error, error, message, and related fields exactly as returned.
  • Never fabricate progress percentages or inferred causes.

Questions people ask

Can it upload a document and parse it in one step?
No. Uploading does not start parsing; parsing begins only when explicitly requested, and its progress is checked separately with the parse-status script.
How does it handle dataset or document deletion?
It first resolves explicit dataset or document IDs and lists the exact items. Deletion proceeds only after the user gives explicit confirmation.
Can retrieval be limited to specific data?
Yes. Searches can target one or more dataset IDs and can also be restricted by document IDs. The retrieval-test endpoint is reserved for single-dataset debugging or explicit requests.

Related skills

Java

Write, debug, and tune Java and JVM systems with JDK-aware code and diagnostic steps.

by Iván130 installs9 stars

minara

Execute crypto research, wallet actions, swaps, perps, and limit orders through the Minara CLI.

162 installs109 stars

neural-memory

Persist and recall related agent memories through a local neural graph with causal and temporal links.

419 installs10 stars

AWS | Amazon Web Services

Architect, troubleshoot, secure, and cost-control AWS infrastructure with explicit cost and blast-radius guidance.

by Iván138 installs2 stars

JSON

Diagnose, validate, transform, and evolve JSON payloads across parsers, schemas, storage, and large files.

by Iván112 installs3 stars

Translate

Translate source material for a named locale while preserving meaning, terminology, placeholders, and locale rules.

380 installs10 stars