Skip to main content
skill CX Codex

recall · parcadei — Claude Code skill: search memory from past sessions

Pulls learnings from past sessions out of the memory system: patterns, errors, installs. Useful when the agent retraces ground you already covered.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add parcadei/Continuous-Claude-v3 --skill recall -a claude-code
Open on GitHub

What recall does

recall runs a semantic search over stored learnings from earlier sessions. You get the top five hits with full content, the learning type, a confidence level and the session they came from.

Storage is PostgreSQL with BGE embeddings. The skill shells out to scripts/core/recall_learnings.py through uv, so it expects a $CLAUDE_OPC_DIR directory holding that script and a reachable database. Miss either and the call fails, which is worth checking before you lean on it mid-task.

Add --k N to change how many results come back, --vector-only for pure vector search, or --text-only when you want speed over precision. The skill is marked as not user-invocable, so the agent is what triggers it.

What it helps with

  • find hook development patterns from earlier sessions
  • recall how the wizard installation went
  • pull up context on TypeScript errors
  • check what has already been tried in this project
  • read the full text of a learning, not just its title

Example prompts for the agent

  • /recall hook development patterns
  • /recall wizard installation
  • /recall TypeScript errors --k 10 --vector-only

How to install recall

1Skills CLI (Vercel)

npx skills add parcadei/Continuous-Claude-v3 --skill recall -a claude-code

2Manually: put the files in place

~/.claude/skills/recall/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/recall/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about recall

How do I install recall in Claude?

Run npx skills add parcadei/Continuous-Claude-v3 --skill recall -a claude-code. The skill lands at ~/.claude/skills/recall/SKILL.md.

What does recall do?

It searches the memory system for learnings from past sessions and returns the top five with their full content, type and confidence.

What does recall need to work?

A $CLAUDE_OPC_DIR directory containing scripts/core/recall_learnings.py, uv to run it, and PostgreSQL with BGE embeddings.

Source text

SKILL.md · MIT
---
name: recall
description: Query the memory system for relevant learnings from past sessions
user-invocable: false
---

# Recall - Semantic Memory Retrieval

Query the memory system for relevant learnings from past sessions.

## Usage

```
/recall <query>
```

## Examples

```
/recall hook development patterns
/recall wizard installation
/recall TypeScript errors
```

## What It Does

1. Runs semantic search against stored learnings (PostgreSQL + BGE embeddings)
2. Returns top 5 results with full content
3. Shows learning type, confidence, and session context

## Execution

When this skill is invoked, run:

```bash
cd $CLAUDE_OPC_DIR && PYTHONPATH=. uv run python scripts/core/recall_learnings.py --query "<ARGS>" --k 5
```

Where `<ARGS>` is the query provided by the user.

## Output Format

Present results as:

```
## Memory Recall: "<query>"

### 1. [TYPE] (confidence: high, id: abc123)
<full content>

### 2. [TYPE] (confidence: medium, id: def456)
<full content>
```

## Options

The user can specify options after the query:

- `--k N` - Return N results (default: 5)
- `--vector-only` - Use pure vector search (higher precision)
- `--text-only` - Use text search only (faster)

Example: `/recall hook patterns --k 10 --vector-only`

© parcadei · MIT · parcadei/Continuous-Claude-v3

Similar skills

All skills for Agent setup: memory, context, orchestration →

recall by other authors comes first.

recall rohitg00 Searches past observations, sessions, and learnings through agentmemory. Use it to recover context when you need to know what a project already covered. CX Codex +1 13,900installs recall backnotprop Rebuilds a short brief from your chat transcripts, the live repo state and the shared record: where things stand and what to do next. CX Codex +1 5,000installs recall michael-denyer Rebuilds your recent working context from chat history, live repo state and the shared project record, then hands back a short brief on where things stand. CX Codex +1 833installs recall arjunkmrm Full-text search across past Claude Code, Codex, pi and Grok sessions with BM25 ranking. Useful when you need to remember what you already discussed with an agent. CX Codex +1 579installs recall entireio Finds previous work sessions via the Entire CLI and turns the closest checkpoint into an actionable playbook. CX Codex +1 404installs pgvector-semantic-search tigerdata Configures AI agents to set up vector similarity search in PostgreSQL using pgvector. Helps store embeddings with halfvec, build HNSW indexes, and power RAG apps. CX Codex +1 938installs vector-database-engineer Seth Hobson A subagent for building vector search: picking a database, choosing embeddings, tuning indexes and hybrid retrieval for RAG. CX Codex ★ 41,729repo stars rag-engineer sickn33 A skill for designing retrieval-augmented generation systems with vector databases. Covers semantic chunking, embedding model selection, and search quality optimization before results reach the LLM. CX Codex +1 1,000installs rag-retrieval OrchestKit A skill with ready-made retrieval-augmented generation patterns: hybrid search, HyDE, reranking, pgvector. Useful when you are wiring a RAG pipeline and want working recipes. CX Codex +1 1,500installs embedding-strategies Seth Hobson This skill helps choose embedding models, tune chunking, and assess quality for semantic search and RAG projects. CX Codex 11,391installs rag-architect https://github.com/Jeffallan Designs production RAG systems: chunks documents, generates embeddings, configures vector stores, builds hybrid search pipelines, and evaluates retrieval quality. CX Codex +1 4,700installs

From the same set69

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor