recall

Query a local associative-memory index for cited workspace answers.

10|1|Updated Jun 29, 2026

Search Tags:#indexing #knowledge management #recall #semantic search #retrieval #memory

Ready to run

npx skills add https://github.com/mishahanin/heading-os --skill recall-mishahanin

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: recall
Source: https://github.com/mishahanin/heading-os/tree/main/.claude/skills/recall
Command: npx skills add https://github.com/mishahanin/heading-os --skill recall-mishahanin

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python3, and includes scripts (resource) components.

What problem does it solve?

This skill solves the problem of fragmented information by providing a unified, semantic search interface across your entire workspace, ensuring past decisions, notes, and context are never lost.

Core Features & Use Cases

Semantic Retrieval: Uses hybrid dense and BM25 search to find information by meaning rather than just keywords, supporting cross-lingual queries.
Cited Answers: Generates responses grounded strictly in retrieved file content with direct path citations, preventing hallucinations.
Use Case: When you need to verify a past decision regarding a specific client engagement or recall a technical detail from a previous project thread, this skill surfaces the exact source files instantly.

Quick Start

Use the recall skill to search your workspace memory for what we decided about the project timeline.

Frequently Asked Questions about recall

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I perform semantic search across my workspace to find past decisions and notes?▼

Semantic search across your workspace is performed using a local associative-memory index that combines hybrid dense and BM25 retrieval to surface past decisions and notes. This approach finds information by meaning rather than exact keywords.

Can I retrieve information from CRM threads and knowledge bases without incurring API costs?▼

Yes, you can retrieve information from CRM threads and knowledge bases with zero API cost. The system operates entirely locally, querying a local associative-memory index to provide context-aware answers while ensuring data privacy.

Does semantic memory retrieval work with cross-lingual queries?▼

Semantic memory retrieval supports cross-lingual queries by utilizing hybrid dense and BM25 search mechanisms. This allows the system to find relevant information based on meaning rather than being restricted by language-specific keywords.

How do I verify the accuracy of cited answers to prevent hallucinations?▼

To prevent hallucinations, cited answers are generated strictly from retrieved file content and include direct path citations. This grounds the responses in actual workspace data, allowing you to verify the exact source files instantly.

What is the best way to index fragmented information for unified knowledge management?▼

The best way to index fragmented information for unified knowledge management is by using a local associative-memory index. This creates a unified, semantic search interface across diverse data layers, ensuring context is never lost.

Do I need Python to run local BM25 and dense retrieval for memory indexing?▼

Yes, you need Python 3 installed in your environment to run the local BM25 and dense retrieval scripts. This dependency is required to execute the associative-memory indexing and querying mechanisms.