knowledge-retrieval

Semantic search over ingested documents using RAG (LlamaIndex/ChromaDB or Foundational RAG)

Quality

75%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./examples/nvidia-deep-researcher/skills/knowledge-retrieval/SKILL.md

Quality

Content

100%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a tight, well-structured overview for a simple tool-wrapper skill: it gives the exact function to call, the expected result format, and constraints without padding. It appropriately assumes Claude's intelligence.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence — it never explains what RAG or a vector store is, and every section earns its tokens.	3 / 3
Actionability	Gives the concrete tool name `knowledge_retrieval`, a clear 4-step call sequence, and a copy-ready citation format; for an instruction-only tool skill this is concrete and actionable.	3 / 3
Workflow Clarity	The numbered How-to-Use sequence is unambiguous for a simple read-only search skill; no destructive or batch operations exist that would require validation checkpoints.	3 / 3
Progressive Disclosure	Under 50 lines with no need for external references, organized into clearly labeled sections (When to Use, How to Use, Result Format, Constraints, Backend Options).	3 / 3
	Total	12 / 12 Passed

Description

50%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description conveys the core capability and backing tech but lacks an explicit "Use when" trigger and uses fairly technical terminology. It is competent but leaves Claude without clear activation guidance.

Suggestions

Add an explicit trigger clause, e.g. "Use when searching pre-ingested documents, PDFs, whitepapers, or domain knowledge not on the open web."

Broaden trigger terms to include natural user phrasings like "knowledge base", "internal documents", and "search the docs".

Name more concrete actions (e.g. "retrieve cited passages", "pull ranked text chunks with filename and page citations") to lift specificity.

Dimension	Reasoning	Score
Specificity	Names the domain and tooling ("semantic search over ingested documents using RAG", "LlamaIndex/ChromaDB") but lists only one concrete action (semantic search) rather than multiple specific operations.	2 / 3
Completeness	Clearly states what the skill does but has no "Use when..." clause or equivalent explicit trigger guidance, which per the rubric caps completeness at 2.	2 / 3
Trigger Term Quality	Includes relevant terms ("semantic search", "ingested documents", "RAG") but leans technical; misses common natural phrasings like "search the knowledge base" or "search documents/PDFs".	2 / 3
Distinctiveness Conflict Risk	The RAG-over-ingested-documents niche is somewhat distinct but could overlap with general document/web/paper search skills without an explicit scoping trigger.	2 / 3
	Total	8 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: open-gitagent/opengap
Path: examples/nvidia-deep-researcher/skills/knowledge-retrieval/SKILL.md
Commit: d7a8e2e

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.