feat: add haystack — expert skill for production search and NLP pipelines #80

New issue

Closed

opened 2026-07-09 14:40:25 -04:00 by jasper · 1 comment

jasper commented

2026-07-09 14:40:25 -04:00

Contributor

Summary

Add a haystack expert skill for production NLP search pipelines. Haystack (by deepset) is the most mature open-source framework for search-specific workloads — pipeline DAG composition, document stores, retrievers, readers, and evaluation — and is already referenced in both the LlamaIndex and LangChain Framework Routing Guides as the recommendation for search pipelines.

Why This Skill

The routing guides in both llamaindex and langchain already say "Production NLP search pipelines → Haystack." Currently that points to a framework we have no skill for. Closing this gap makes the routing portfolio complete for the search use case.

Scope

Deep research: Haystack architecture (Pipelines, DocumentStore, Retriever, Reader, Generator), v2.x API, custom components, indexing vs query pipelines, evaluation with deepset-eval
Vault notes: atoms + molecules on Haystack pipeline model, document processing, retrieval-augmented generation patterns
Expert skill: SKILL.md + references (pipeline design, document stores, retrievers, evaluation, deployment), templates (indexing, query, QA), scripts
Greenfield SkillOpt: 3 epochs
PR: Single PR

Key References

https://docs.haystack.deepset.ai — official docs
https://github.com/deepset-ai/haystack — GitHub

Acceptance Criteria

Deep research pyramid
Vault clipping + 20+ atoms + 4 molecules
Agent skill with 7+ references, 3+ templates, 1+ script
Greenfield SkillOpt 3 epochs
PR merged to main

## Summary Add a **haystack** expert skill for production NLP search pipelines. Haystack (by deepset) is the most mature open-source framework for search-specific workloads — pipeline DAG composition, document stores, retrievers, readers, and evaluation — and is already referenced in both the LlamaIndex and LangChain Framework Routing Guides as the recommendation for search pipelines. ## Why This Skill The routing guides in both llamaindex and langchain already say "Production NLP search pipelines → Haystack." Currently that points to a framework we have no skill for. Closing this gap makes the routing portfolio complete for the search use case. ## Scope - **Deep research:** Haystack architecture (Pipelines, DocumentStore, Retriever, Reader, Generator), v2.x API, custom components, indexing vs query pipelines, evaluation with deepset-eval - **Vault notes:** atoms + molecules on Haystack pipeline model, document processing, retrieval-augmented generation patterns - **Expert skill:** SKILL.md + references (pipeline design, document stores, retrievers, evaluation, deployment), templates (indexing, query, QA), scripts - **Greenfield SkillOpt:** 3 epochs - **PR:** Single PR ## Key References - https://docs.haystack.deepset.ai — official docs - https://github.com/deepset-ai/haystack — GitHub ## Acceptance Criteria - [ ] Deep research pyramid - [ ] Vault clipping + 20+ atoms + 4 molecules - [ ] Agent skill with 7+ references, 3+ templates, 1+ script - [ ] Greenfield SkillOpt 3 epochs - [ ] PR merged to main

jasper commented

2026-07-11 16:49:40 -04:00

Author

Contributor

Resolved by merged PR #84: #84

Filed by Jasper (AI agent on behalf of Magnus Hedemark)

Resolved by merged PR #84: https://git.brandyapple.com/magnus/agent-skills/pulls/84 Filed by Jasper (AI agent on behalf of Magnus Hedemark)

jasper closed this issue

2026-07-11 16:49:41 -04:00

No labels

community-feedback

enhancement

skill-upgrade

No milestone

No project

No assignees

1 participant

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

magnus/agent-skills#80

No description provided.

Rows
Columns