Skill Claude CodeCodex
Build a trustworthy first ontology from an empty or near-empty ontology-atlas vault using only Atlas MCP evidence. Use when the user asks to analyze a codebase, bootstrap/fill its ontology, extract product meaning from a repository, or when a requested ontology task finds only starter nodes. Separate observed…
Skill Claude CodeCodex
Extract a small, evidence-bound set of ontology candidates from prose, check the existing vault for duplicates, obtain user approval, and land only the approved nodes and relations.
Skill Claude CodeCodex
Measure whether Atlas actually works on a repository nobody here knows, by building a vault with a real MCP agent and then handing that vault — without the source — to a second agent who must answer questions from it alone. Use when changing the construction rules, the MCP read/write contract, the bootstrap skill, or…
Skill Claude CodeCodex
After a code change, sync the project's ontology vault — read what's already there, identify new capabilities / elements / domains introduced by the change, and write them back via the MCP server (or fall back to the CLI). Use this at the end of any task that introduces a new feature, refactors a module, or renames a…
Skill Claude CodeCodex
Write a subagent brief that isolates ports, files, worktrees, scratch output, baselines, and primary evidence before parallel work begins.
Skill Claude CodeCodex
Run five independent product-owner seats, one rebuttal round, and one accountable decision with dissent and a falsifier for expensive or hard-to-reverse work.
Skill Claude CodeCodex
Run the solo product-owner gate before product, UX, graph, MCP, CLI, workflow, or macOS-shell work. Separate phenomenon from problem, score all six rubric rows, and escalate mechanically.
Skill Claude CodeCodex
Live-verify layout across tablet, laptop, and wide breakpoints by measuring rects, elementFromPoint occlusion, bottom-tab reserve, and screenshots.
Skill Claude CodeCodex
Walk one complete journey against the running build with a declared knowledge state, name observable UX failure patterns, and refuse to invent whether a person would want the product.
Hook Claude Code
Runs before the agent uses a tool for Bash, Edit, Write, MultiEdit and NotebookEdit tool calls, executing block-npm-publish.sh, block-unsafe-git.sh and block-generated-edit.sh (3 commands). From wlsdks/ontology-atlas.
Hook Claude Code
Runs after a tool call finishes for Edit, Write, MultiEdit and NotebookEdit tool calls, executing report-agent-file-drift.sh. From wlsdks/ontology-atlas.
Hook Claude Code
Runs when a session starts, executing inject-ontology-summary.sh. From wlsdks/ontology-atlas.
Settings file Claude Code
Agent settings declaring 3 hook events (PreToolUse, PostToolUse, SessionStart).
MCP server Claude CodeCodexCursor +2
MCP server "ontology-atlas" as configured in wlsdks/ontology-atlas. Runs ./mcp/src/index.js with node. Needs 2 environment variables to run.
MCP server Claude CodeCodexCursor +2
Lets the agent inspect and automate a live Chrome instance through the DevTools protocol: performance traces, network, console. Runs locally from the chrome-devtools-mcp npm package.
Instructions file CodexOpenCode
Instructions for wlsdks/ontology-atlas, covering agents.md — ontology-atlas, product and non-negotiable architecture, start here, structure and routes and operating gates and skills.
Instructions file
Instructions for wlsdks/ontology-atlas, covering claude.md, visibility and mirrors, claude code loading and synchronization.
Skill Claude CodeCodex
A document-reading skill that finds concepts in meeting notes, plans, or pull-request descriptions and connects them to an existing knowledge graph. A knowledge graph stores concepts and their relationships.
Skill Claude CodeCodex
A Korean-language planning procedure for deciding what knowledge or structure should be added to a vault. A vault here is a managed collection of linked concepts, evidence, and project information.
Skill Claude CodeCodex
A read-only health check for a project knowledge vault, where notes and their links form a map of concepts and components. It explains broken or disconnected parts in ordinary language.
Agent
Ontology Atlas is a single-context product implemented across the web workbench, the macOS shell, the CLI, and the MCP server. Engineering skills should use the same product vocabulary across all of these packages.
Agent
Use GitHub Issues for work that will not be completed immediately, needs discussion or reproduction evidence, or spans multiple implementation slices. Small, well-defined changes may proceed directly from a branch to a pull request without creating an issue.
Agent
Engineering skills use five canonical triage roles. Map each role to the same label name in GitHub Issues.