Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →shadd0wtaka/rag-vector-bases-skill
Workflows für Retrieval-Augmented Generation, Embeddings, Vector-DBs, Chunking-Strategien.
Passed 20 of 22 test cases with the skill, and 17 of 22 without it.
and 3 more in the full report
3 fixed17 passed either way2 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 3 test cases the skill fixed. Same model, same prompt, with and without the skill.
I need a Python script using LangChain to split markdown files recursively. It should split text using natural boundaries like headers and paragraphs while preserving metadata such as the source filename and page number. Please provide the splitter configuration and document creation code.
| Checked | Without skill | With skill |
|---|---|---|
| Instantiates RecursiveCharacterTextSplitter | Passed | Passed |
| Configures separators array including double hash and triple hash markdown headers | Missed | Passed |
| Calls create_documents with metadatas parameter | Missed | Passed |
The instructions your model reads when the skill loads.
Workflows für Retrieval-Augmented Generation, Embeddings, Vector-DBs, Chunking-Strategien.
def fixed_chunks(text: str, chunk_size: int = 512, overlap: int = 64) -> list[str]:
chunks = []
start = 0
while start < len(text):
end = start + chunk_size
chunks.append(text[start:end])
start += chunk_size - overlap
return chunks
from langchain.text_splitter import RecursiveCharacterTextSplitter
splitter = RecursiveCharacterTextSplitter(
chunk_size=1000,
chunk_overlap=200,
separators=["\n## ", "\n### ", "\n\n", "\n", ". ", " ", ""],
)
chunks = splitter.split_text(text)
# Mit Metadaten
docs = splitter.create_documents(texts=[text], metadatas=[{"source": "file.md", "page": 1}])
import httpx
resp = httpx.post(
"http://localhost:20128/v1/embeddings",
test cases passed, vs 17 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.