home / skills / nilecui / skillsbase / remembering-conversations

remembering-conversations skill

safe

/.cursor/skills/remembering-conversations

This skill helps you quickly locate past conversations using semantic or exact text search to surface decisions, patterns, and context.

npx playbooks add skill nilecui/skillsbase --skill remembering-conversations

Review the files below or copy the command above to add this skill to your agents.

Files (29)

SKILL.md

2.0 KB

---
name: Remembering Conversations
description: Search previous Claude Code conversations for facts, patterns, decisions, and context using semantic or text search
when_to_use: when partner mentions past discussions, debugging familiar issues, or seeking historical context about decisions and patterns
version: 1.1.0
---

# Remembering Conversations

Search archived conversations using semantic similarity or exact text matching.

**Core principle:** Search before reinventing.

**Announce:** "I'm searching previous conversations for [topic]."

**Setup:** See INDEXING.md

## When to Use

**Search when:**
- Your human partner mentions "we discussed this before"
- Debugging similar issues
- Looking for architectural decisions or patterns
- Before implementing something familiar

**Don't search when:**
- Info in current conversation
- Question about current codebase (use Grep/Read)

## In-Session Use

**Always use subagents** (50-100x context savings). See skills/using-skills for workflow.

**Manual/CLI use:** Direct search (below) for humans outside Claude Code sessions.

## Direct Search (Manual/CLI)

**Tool:** `${SUPERPOWERS_SKILLS_ROOT}/skills/collaboration/remembering-conversations/tool/search-conversations`

**Modes:**
```bash
search-conversations "query"              # Vector similarity (default)
search-conversations --text "exact"       # Exact string match
search-conversations --both "query"       # Both modes
```

**Flags:**
```bash
--after YYYY-MM-DD    # Filter by date
--before YYYY-MM-DD   # Filter by date
--limit N             # Max results (default: 10)
--help                # Full usage
```

**Examples:**
```bash
# Semantic search
search-conversations "React Router authentication errors"

# Find git SHA
search-conversations --text "a1b2c3d4"

# Time range
search-conversations --after 2025-09-01 "refactoring"
```

Returns: project, date, conversation summary, matched exchange, similarity %, file path.

**For details:** Run `search-conversations --help`

Overview

This skill helps you locate facts, patterns, decisions, and context from past Claude Code conversations using semantic similarity or exact text matching. It speeds decision-making by reusing prior discussions and avoiding repeated investigation. When invoked, the agent announces: "I'm searching previous conversations for [topic]."

How this skill works

The tool performs vector-based semantic search by default and can switch to exact text matching or run both modes. You can filter results by date range and limit the number of hits; each result shows project, date, a summary, a matched exchange, similarity score, and file path. For interactive sessions, it runs as a subagent to save context; a standalone CLI is available for manual queries.

When to use it

When someone says "we discussed this before" and you want the original context
When debugging recurring issues or looking for past troubleshooting steps
When searching for architectural decisions, patterns, or accepted trade-offs
Before reimplementing known functionality to avoid duplicating work
When you need to find a specific git SHA, snippet, or quoted text from past chats

Best practices

Announce searches to collaborators: "I'm searching previous conversations for [topic]"
Prefer semantic search for concept matches and exact text mode for precise identifiers
Use date filters to narrow results when the conversation window is known
Run searches as subagents within sessions to minimize context costs
Limit results to a manageable number and inspect summaries before full reads

Example use cases

Find prior discussion about React Router authentication failures to reuse fixes
Locate an earlier decision on database sharding and the reasoning behind it
Search for a previously shared git SHA or stack trace using exact text mode
Retrieve a past design pattern or code suggestion to apply to a new feature
Quickly gather prior troubleshooting steps when a bug reappears after a refactor

FAQ

What search modes are available?

Vector semantic search is the default; there is also exact text matching and a combined mode that runs both.

Can I filter results by time?

Yes — you can filter with after and before date ranges to narrow the returned conversations.

How are results presented?

Each hit includes project, date, a conversation summary, the matched exchange, a similarity percentage, and the file path.