search-similar

search-similar is a command for Claude Code from rohitg00/awesome-claude-code-toolkit. It costs 0 tokens per session (281 once invoked), scanned A, original, Apache-2.0.

A command for finding text or records that have similar meaning in a vector database. It ranks results by similarity and can filter, re-rank, and remove near-duplicates.

In plain words
What is it for?
Use it to search indexed documents, database records, or other text by meaning, with optional category, date, or source filters.
Why use it?
It helps when exact keyword searches miss relevant items or return too many loosely related results. Similarity scores and source details make the matches easier to assess.

Command for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the embedding-manager plugin — 2 commands shipped together

Good fit Use it to search indexed documents, database records, or other text by meaning, with optional category, date, or source filters.

Compare 6 commands from other repositories ↓
Install with agentmods
npx agentmods add commands/rohitg00/awesome-claude-code-toolkit/search-similar
About the project

Awesome Claude Code Toolkit is a curated collection of extensions and configuration for Claude Code, including agents, skills, commands, plugins, hooks, rules, templates, MCP configurations, and companion apps. It is for Claude Code users who want ready-made workflows and integrations for different development tasks. The catalogue add-ons are selected components from this toolkit.

rohitg00/awesome-claude-code-toolkit · 2,598 stars · on GitHub

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.

Clone the repo
git clone --depth 1 https://github.com/rohitg00/awesome-claude-code-toolkit

Made for: Claude Code.

Or install embedding-manager, the plugin that ships this one along with the rest of its 2 commands.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for search-similar

README.md
[![agentmods](https://agentmods.dev/badge/commands/rohitg00/awesome-claude-code-toolkit/search-similar.svg)](https://agentmods.dev/commands/rohitg00/awesome-claude-code-toolkit/search-similar)
Your own site
<a href="https://agentmods.dev/commands/rohitg00/awesome-claude-code-toolkit/search-similar"><img src="https://agentmods.dev/badge/commands/rohitg00/awesome-claude-code-toolkit/search-similar.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 281 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00000 $0.00281
Opus 5 $0.00000 $0.00140
Sonnet 5 $0.00000 $0.00056
Haiku 4.5 $0.00000 $0.00028

Measured 4d ago against content hash 9a0f2c6aac15, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

search-similar 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 4d ago.

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/embedding-manager/commands/search-similar.md · 29 lines

What it actually says

/search-similar - Search Similar Items

Find semantically similar items using vector similarity search.

Steps

  1. Take the search query text from the user
  2. Generate an embedding for the query using the same model as the index
  3. Perform approximate nearest neighbor search in the vector store
  4. Retrieve the top-K most similar items (default: 10)
  5. Calculate and display similarity scores for each result
  6. Apply metadata filters if specified (category, date range, source)
  7. Re-rank results using cross-encoder for improved precision
  8. Deduplicate results that are too similar to each other (similarity > 0.95)
  9. Format results with: rank, similarity score, source text, metadata
  10. Provide a relevance assessment for the top results
  11. Suggest query refinements if results are not satisfactory
  12. Cache the query embedding for repeated searches

Rules

  • Use the same embedding model for queries and indexed items
  • Set a minimum similarity threshold to filter irrelevant results (default: 0.5)
  • Return metadata with results for context and source attribution
  • Handle empty results gracefully with alternative search suggestions
  • Limit result count to avoid overwhelming output (max 50)
  • Include the similarity metric used (cosine, dot product, euclidean)
  • Cache frequently searched queries for faster response times
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. 4d ago First seen · 29 lines · 0 tokens per session scan A 9a0f2c6aac15

Subscribe to this mod's changes

search-similar is a command published in the GitHub repository rohitg00/awesome-claude-code-toolkit (2,598 stars, last pushed 3mo ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 281 tokens. 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 2026-09-03.