connection-finder

A specialist agent that searches a personal knowledge graph for less obvious links between notes. A knowledge graph is a connected collection of notes and their relationships.

In plain words
What is it for?
Use it to search permanent notes, extracted insights, documents, source material, maps, and articles, then record discovered connections and session changes.
Why use it?
It helps find relationships and patterns that are easy to miss when reading notes one at a time.

Agent for Claude Code

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 agents/abilityai/cornelius/connection-finder
Clone the repo
git clone --depth 1 https://github.com/Abilityai/cornelius

Made for: Claude Code.

Per session 31 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 5,403 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 $0.00031 $0.05403
Opus 5 $0.00015 $0.02701
Sonnet 5 $0.00006 $0.01081
Haiku 4.5 $0.00003 $0.00540

Measured 2d ago against content hash 02a4d99ffe20, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

connection-finder 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 2d 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.

.claude/agents/connection-finder.md · 677 lines

How it starts

The opening of the file, as written. The whole thing — 677 lines — stays where its author put it; the contents beside it link to each section on GitHub.

State Dependencies

Source Location Read Write Description
Permanent Notes Brain/02-Permanent/ Primary connection hunting ground
AI Extracted Notes Brain/AI Extracted Notes/ AI-harvested insights
Document Insights Brain/Document Insights/ External document insights
Source Materials Brain/01-Sources/ Trace influence
MOCs Brain/03-MOCs/ Existing navigation maps
Articles Brain/04-Output/Articles/ Existing synthesis
Local Brain Search resources/local-brain-search/ Semantic search and connections
Session Changelogs Brain/05-Meta/Changelogs/ Dated session changelog
Master Changelog Brain/CHANGELOG.md Brief summary entry

Use Local Brain Search for all semantic search and connection discovery.

Location: resources/local-brain-search/

Wrapper Scripts:

# Semantic search
resources/local-brain-search/run_search.sh "query" --limit 10 --json

# Find connections for a note
resources/local-brain-search/run_connections.sh "Note Name" --json

# Get hub notes
resources/local-brain-search/run_connections.sh --hubs --json

# Get graph statistics
resources/local-brain-search/run_connections.sh --stats --json

# Find bridge notes
resources/local-brain-search/run_connections.sh --bridges --json

READ SCOPE: with scope enforcement on, a bare command reads only the core fingerprint (02-Permanent, 03-MOCs, AI Extracted Notes, 01-Sources). That is the right default when you are mapping the curated graph (e.g. the /find-connections command). But when your task is to connect newly-created non-core notes into the graph - e.g. spawned by deep-research / economist-analyze to map connections for fresh Document Insights/ notes, or by ingest-source for fresh Books/<slug>/ notes - those targets live outside core, so a bare command can neither resolve nor reach them. In that case prefix every search/connection command with the scopes your task spans, e.g.:

Read the full file on GitHub · 677 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. 2d ago First seen · 677 lines · 31 tokens per session scan A 02a4d99ffe20

Subscribe to this mod's changes

connection-finder is an agent published in the GitHub repository Abilityai/cornelius (104 stars, last pushed 9d ago), licensed MIT. It adds 31 tokens to every session and 5,403 once invoked, about $0.0002 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.