Skip to main content

Living knowledge map for codebases — install agent instructions across AI platforms

Project description

kb-builder

Kode Brain — a living knowledge system for evolving software projects.

Converts an imperfect, growing codebase into a structured, searchable knowledge map of domains, capabilities, concepts, flows, runtime behavior, dependencies, legacy areas, migration states, and source evidence — so humans and AI agents can understand and modify the system without rediscovering everything from scratch.

Design Documents

Document Purpose
Taxonomy Finalized node types, edge types, status labels, confidence labels
Skills Skill API contracts — inputs, outputs, guarantees
Agents Agent role boundaries — responsibilities, allowed skills, forbidden actions
Workflows Core workflow sequence diagrams
Open Decisions Unresolved architectural decisions

Schemas

File Purpose
schema/node.schema.json JSON Schema for KnowledgeNode
schema/edge.schema.json JSON Schema for KnowledgeEdge
schema/knowledge-base.schema.json Top-level graph container schema

Design Order

Per the Kode Brain spec (§20.7):

Taxonomy → Workflow → Skills → Agents → Plugin/CLI

This repository is currently at the design phase. Implementation comes after the design is validated.

Commands

/kodebrain init [path]                     Scan project, scaffold docs/kodebrain/
/kodebrain scan [path]                     Re-scan and update knowledge graph
/kodebrain query "<task or symptom>"       Query the KB by task description
/kodebrain reading-pack "<task>"           Generate + save a context pack
/kodebrain detect-legacy [--domain slug]   Flag suspected dead code
/kodebrain review [--page path]            Review KB pages for stale claims
/kodebrain update [--diff] [--files ...]   Update KB from changed files

Installation

The skill must be installed into Claude Code's skills directory:

# From the repo root
ln -s "$(pwd)" ~/.claude/skills/kodebrain

After symlinking, SKILL.md is discoverable by Claude Code and /kodebrain becomes available in any Claude Code session.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

kodebrain-0.4.0.tar.gz (60.0 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

kodebrain-0.4.0-py3-none-any.whl (26.4 kB view details)

Uploaded Python 3

File details

Details for the file kodebrain-0.4.0.tar.gz.

File metadata

  • Download URL: kodebrain-0.4.0.tar.gz
  • Upload date:
  • Size: 60.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.4

File hashes

Hashes for kodebrain-0.4.0.tar.gz
Algorithm Hash digest
SHA256 b658036ba9864987fa5a8de9551483d425b8cf239616554bf1bedbd621dea1f6
MD5 b59cf26875e6ba08fe0cb8d288a998c6
BLAKE2b-256 fc2c21f4c663174e771d30f5c7a3229bb9c170ae9091bd56f47b42cd77386f74

See more details on using hashes here.

File details

Details for the file kodebrain-0.4.0-py3-none-any.whl.

File metadata

  • Download URL: kodebrain-0.4.0-py3-none-any.whl
  • Upload date:
  • Size: 26.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.4

File hashes

Hashes for kodebrain-0.4.0-py3-none-any.whl
Algorithm Hash digest
SHA256 59a1502405e3a41731043fcb291ef1302d15d08d214c8ce976ca6125fb781866
MD5 815970136adab92639f704d936436534
BLAKE2b-256 eff5b2dd7c477aedf818c3579b8d85f7c09c0e792fb4c2636344b42bee79576e

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page