スキル一覧に戻る
mindmorass

rag-wrapper

by mindmorass

0🍴 0📅 2026年1月19日
GitHubで見るManusで実行

SKILL.md


name: rag-wrapper description: Patterns for wrapping any agent with RAG context from Qdrant. Use to add persistent memory to imported or external agents.

RAG Wrapper Patterns

Patterns for augmenting any agent with Qdrant context retrieval.

Quick Start

To wrap an agent with RAG:

Use rag-proxy agent:
  Target: {agent-to-wrap}
  Task: {the task}

Manual Wrapping Pattern

If you need custom control, follow this pattern:

Step 1: Query Relevant Context

Tool: qdrant-find
Query: {key terms from task}

Step 2: Format Context Block

## Retrieved Context

### Source: {metadata.source}
Harvested: {metadata.harvested_at}
Type: {metadata.type}

{document content}

---

Step 3: Prepend to Task

{context blocks}

## Task

{original task}

---
Note: Above context is from stored knowledge. Verify if needed.

Step 4: Delegate

Tool: Task
Agent: {target-agent}
Prompt: {enriched prompt}

Enriched Prompt Template

# Context from Stored Knowledge

The following relevant information was retrieved from project memory:

{{#each contexts}}
## From Qdrant
**Source:** {{metadata.source}}
**Harvested:** {{metadata.harvested_at}}

{{content}}

---
{{/each}}

# Your Task

{{original_task}}

---

**Note:** The context above comes from previously harvested research.
Use it if relevant, but verify currency for time-sensitive information.
The `harvested_at` dates indicate when the content was stored.

Selective Wrapping

Not all tasks need RAG. Skip for:

Task TypeWrap?Reason
Fresh researchNoNeed current, not cached data
Simple editsNoContext not needed
RAG-aware agentsNoAlready query Qdrant
ImplementationYesBenefit from patterns, decisions
DebuggingYesPrevious solutions may help
ArchitectureYesDecisions and constraints matter

Agent-Collection Affinity

Map agent types to useful query topics:

Agent TypeQuery Topics
frontend-developerreact, design system, components
backend-architectapi, architecture, decisions
security-auditorsecurity, authentication, vulnerabilities
devopsinfrastructure, terraform, deployment
testertesting, coverage, quality

Storing Results

After the target agent completes:

Tool: qdrant-store
Information: "<valuable findings>"
Metadata:
  source: "agent-output"
  type: "generated"
  harvested_at: "<ISO date>"
  tags: "<relevant,keywords>"

Error Handling

ScenarioAction
Empty query resultsProceed without context
Qdrant unavailableFall back to unwrapped delegation
Target agent failsReport error, don't retry with less context

スコア

総合スコア

50/100

リポジトリの品質指標に基づく評価

SKILL.md

SKILL.mdファイルが含まれている

+20
LICENSE

ライセンスが設定されている

0/10
説明文

100文字以上の説明がある

0/10
人気

GitHub Stars 100以上

0/15
最近の活動

3ヶ月以内に更新がある

0/10
フォーク

10回以上フォークされている

0/5
Issue管理

オープンIssueが50未満

+5
言語

プログラミング言語が設定されている

+5
タグ

1つ以上のタグが設定されている

0/5

レビュー

💬

レビュー機能は近日公開予定です