search

A command that searches Claude Code's saved session history for earlier decisions, discussions, summaries, and tool results.

In plain words
What is it for?
Finding past decisions about authentication, database design, bug fixes, or other project topics, with options to search summaries or recent sessions.
Why use it?
It helps recover useful context when an earlier conversation contains information needed for current work.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/stefan-jansen/claude-code-toolkit/search
Clone the repo
git clone --depth 1 https://github.com/stefan-jansen/claude-code-toolkit
Per session 11 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 965 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00011 $0.00965
Opus 5 $0.00005 $0.00483
Sonnet 5 $0.00002 $0.00193
Haiku 4.5 $0.00001 $0.00097

Measured yesterday against content hash 9d987c198847, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

search scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/memory/commands/search.md · 139 lines

How it starts

The opening of the file, as written. The whole thing — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/memory:search [query]

Search Claude Code's session history to find past decisions, discussions, and context.

How It Works

Claude Code automatically stores complete session history in ~/.claude/projects/. This command searches that history to recover:

  • Past decisions and their rationale
  • Discussions about specific topics
  • Auto-compact summaries
  • Tool calls and their results

Usage

/memory:search "authentication approach"
/memory:search "database schema" --summaries-only
/memory:search "bug fix" --recent 5

Arguments

  • query - Keywords to search for (required)
  • --summaries-only - Only search auto-compact summaries (faster, less detail)
  • --recent N - Only search N most recent sessions (default: 20)

Implementation

When invoked, execute:

#!/bin/bash
QUERY="$1"
PROJECT_DIR=$(pwd)
CLAUDE_PROJECT="-$(echo "$PROJECT_DIR" | sed 's|^/||; s|/|-|g')"
SESSION_DIR="$HOME/.claude/projects/$CLAUDE_PROJECT"

if [ ! -d "$SESSION_DIR" ]; then
    echo "No session history found for this project."
    exit 1
fi

echo "Searching session history for: $QUERY"
echo ""

# Search summaries first (fast overview)
echo "## Relevant Summaries"
for f in $(ls -t "$SESSION_DIR"/*.jsonl | head -20); do
    grep -i "$QUERY" "$f" 2>/dev/null | \
        jq -r 'select(.type=="summary") | .summary' 2>/dev/null
done | sort -u | head -15

echo ""
echo "## Key Discussions"

# Search user messages and assistant responses
for f in $(ls -t "$SESSION_DIR"/*.jsonl | head -10); do
    DATE=$(stat -c %y "$f" | cut -d' ' -f1)

    # Check if file matches
    if grep -qi "$QUERY" "$f"; then
        echo ""
        echo "### Session: $DATE"

        # Extract relevant assistant text (decisions, recommendations)
        grep -i "$QUERY" "$f" | \
            jq -r 'select(.type=="assistant") | .message.content[]? |
                   select(.type=="text") | .text' 2>/dev/null | \
            grep -i "$QUERY" | \
            grep -iE "(decided|chose|recommend|should|will use|verdict|conclusion|the answer|solution)" | \
            head -5
    fi
done

Read the full file on GitHub · 139 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 139 lines · 11 tokens per session scan A 9d987c198847

Subscribe to this mod's changes

search is a command published in the GitHub repository stefan-jansen/claude-code-toolkit (85 stars, last pushed 1mo ago), licensed MIT. It adds 11 tokens to every session and 965 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.