search

search is a command for coding agents from zach-blumenfeld/knowledge-index. It costs 0 tokens per session (3,233 once invoked), scanned A, original, Apache-2.0.

A command for searching a knowledge graph, a connected collection of documents and sections, using full-text queries and optional meaning-based expansion.

In plain words
What is it for?
Use it to search indexed documents, sections, and vaults by text, scope, query syntax, or related terms.
Why use it?
It finds relevant document sections without requiring the agent to read or search every file manually.

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/zach-blumenfeld/knowledge-index/search
Clone the repo
git clone --depth 1 https://github.com/zach-blumenfeld/knowledge-index

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

README.md
[![agentmods](https://agentmods.dev/badge/commands/zach-blumenfeld/knowledge-index/search.svg)](https://agentmods.dev/commands/zach-blumenfeld/knowledge-index/search)
Your own site
<a href="https://agentmods.dev/commands/zach-blumenfeld/knowledge-index/search"><img src="https://agentmods.dev/badge/commands/zach-blumenfeld/knowledge-index/search.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 3,233 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 $0.00000 $0.03233
Opus 5 $0.00000 $0.01617
Sonnet 5 $0.00000 $0.00647
Haiku 4.5 $0.00000 $0.00323

Measured 4d ago against content hash 522bd34ef3ae, 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 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.

docs/commands/search.md · 305 lines

How it starts

The opening of the file, as written. The whole thing — 305 lines — stays where its author put it; the contents beside it link to each section on GitHub.

How ki search works

ki search is full-text retrieval over a vault's knowledge graph. It runs one ranked sweep over the documents and sections in scope and returns the matching slices — uri, title, score, and content — so an agent pulls just what it needs instead of grepping and reading whole files.

This doc is the holistic picture: the index, the query, scope resolution (local vs remote, --under and --vault), result shaping, and the external-stub nuance. For the underlying profile/vault model it builds on, see docs/scoping.md.


1. The index

Indexing builds a single Neo4j full-text index, content_search, over three node labels:

(:Document | :Section | :Vault)  ON EACH  [displayName, content, aliases, description]
  • Standard analyzer — lexical, lowercased, whitespace/punctuation tokenized. It does not understand synonyms or meaning (see Semantic expansion), and it shreds punctuation-heavy strings like URIs (which is why scoping is a Cypher predicate, never a Lucene clause — see §5.4).
  • Fields — a hit can come from a node's title (displayName), its body (content), its wikilink aliases, or a vault/section description.

ki search only ever returns Documents and Sections. Vaults are in the index (so a vault description can match) but are filtered out of results by label.


2. One unified sweep (Documents + Sections)

There is a single search path — one query over both Documents and Sections at once (SEARCH_DOC_SECTION). There is no separate "title search" vs "section search" mode.

Why both by default: in ki's data model a Document's content is just its header/intro (everything above the first heading); the rest of the body lives in its Sections. So searching documents alone misses most of the prose, and searching sections alone misses the intro and single-section notes. Sweeping both is the right default.

Narrow with --types when you have reason to:

ki search "token refresh"                      # documents + sections (default)
ki search "token refresh" --types section      # sections only
ki search "token refresh" --types document     # documents only (intros/titles)

Read the full file on GitHub · 305 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. 4d ago First seen · 305 lines · 0 tokens per session scan A 522bd34ef3ae

Subscribe to this mod's changes

search is a command published in the GitHub repository zach-blumenfeld/knowledge-index (6 stars, last pushed 10d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 3,233 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 2026-08-31.