kg-compose

A procedure for turning documents or pasted notes into a queryable knowledge graph, a database of entities and their relationships. It extracts facts, assigns stable identifiers, and writes them into a named ontology, which defines how the information should be interpreted.

In plain words
What is it for?
Converting notes, documents, transcripts, or research into structured graph data. It can create or use an ontology and follow its preference for literal or inferred relationships.
Why use it?
It turns unstructured material into linked facts that can be searched and reused, while recording and controlling each write.

Skill for Claude CodeCodex

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 skills/cunicopia-dev/knowledge-graph-rdbms/kg-compose
Any agent
npx skills add cunicopia-dev/knowledge-graph-rdbms --skill kg-compose
Clone the repo
git clone --depth 1 https://github.com/cunicopia-dev/knowledge-graph-rdbms

Made for: Claude Code, Codex.

Per session 113 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,615 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 $0.00113 $0.01615
Opus 5 $0.00056 $0.00807
Sonnet 5 $0.00023 $0.00323
Haiku 4.5 $0.00011 $0.00161

Measured 2d ago against content hash a96dc1cc211f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

kg-compose 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/skills/kg-compose/SKILL.md · 116 lines

How it starts

The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.

kg-compose — document → ontology

Turn unstructured source material into structured, queryable graph facts in a named kgrdbms ontology. You are the extraction engine; the ontology supplies the opinion (how aggressive to be), and every write is gated and logged so a wrong call is reversible, not permanent.

The one idea

You are mechanism; the ontology is policy. Don't impose a house style — read the target ontology's stance and honor it. A literal legal-notes ontology and an inferential research-notes ontology get different graphs from the same paragraph, on purpose.

Procedure

0. Resolve the target ontology

  • If the user named one, use it. If not, propose a short kebab-case name from the material and confirm.
  • Check whether it exists: kg --json ontology list. If absent, create it: kg ontology create NAME --stance <literal|inferential> --description "…". If present, do not recreate it — read its existing stance/path from the list output and honor them.

1. Read the ontology's opinion

From the registry entry: stance (free-text extraction guidance — see Stance below), allowed_kinds (if non-empty, prefer those kinds; extract others but flag that they're outside the ontology's allowlist), id_convention (default CURIE prefix:slug). These are the guidance you compose within.

2. Decompose the source into a {nodes, edges} model

  • Nodes = the things. Each: id (a CURIE — see Id rules), kind (a TitleCase type like Person, Company, Method), name (display string), labels (set memberships), properties (JSON facts).
  • Edges = the relationships. Each: from, to, type (UPPER_SNAKE verb like FOUNDED, MADE_WITH, REPORTS_TO), and optional properties (facts about the relationship itself — year, confidence, source).
  • Let the stance govern how far past the literal text you go.

3. Write it (bulk, gated, logged — ONE call, not N)

Write the whole {nodes, edges} model in a single bulk operation. Do not emit dozens of individual upsert calls — use the bulk path:

Read the full file on GitHub · 116 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 · 116 lines · 113 tokens per session scan A a96dc1cc211f

Subscribe to this mod's changes

kg-compose is a skill published in the GitHub repository cunicopia-dev/knowledge-graph-rdbms (1 stars, last pushed 23d ago), licensed MIT. It adds 113 tokens to every session and 1,615 once invoked, about $0.0006 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 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

agent-host-chat-contributions

Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.

microsoft/vscode · 56 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens