@nolrm/contextkit 1.1.0 → 1.1.1

package/README.md

@@ -72,7 +72,7 @@ contextkit install claude   # or: contextkit install (interactive picker)
 ck claude      # creates CLAUDE.md + .claude/rules/ — skips if already up to date
 ck cursor      # creates .cursor/rules/ (scoped .mdc files) — skips if already up to date
 ck copilot     # creates .github/copilot-instructions.md
-ck codex       # creates AGENTS.md
+ck codex       # creates AGENTS.md + .codex/skills/ (24 skills)
 ck opencode    # creates AGENTS.md
 ck gemini      # creates GEMINI.md + .gemini/settings.json
 ck aider       # creates CONVENTIONS.md + .aider/rules.md
@@ -134,10 +134,11 @@ claude "create checkout flow for customer"
 @.contextkit Create checkout flow for customer
 ```
 
-**Codex CLI** — reads `AGENTS.md` automatically
+**Codex CLI** — reads `AGENTS.md` automatically + skills in `.codex/skills/`
 
 ```bash
 codex "create checkout flow for customer"
+# Skills available: $spec, $squad, $squad-auto, $squad-spec, $analyze, $ck, and more
 ```
 
 **OpenCode** — reads `AGENTS.md` automatically
@@ -450,7 +451,7 @@ ck install --force    # regenerate all files, including user-customized standard
 ck claude      # add or refresh Claude integration — skips if already up to date
 ck cursor      # add or refresh Cursor integration — skips if already up to date
 ck copilot     # add GitHub Copilot integration
-ck codex       # add Codex CLI integration (AGENTS.md)
+ck codex       # add Codex CLI integration (AGENTS.md + .codex/skills/)
 ck opencode    # add OpenCode integration (AGENTS.md)
 ck gemini      # add Gemini CLI integration (GEMINI.md)
 ck aider       # add Aider integration (CONVENTIONS.md)

package/lib/integrations/codex-integration.js

@@ -7,17 +7,57 @@ class CodexIntegration extends BaseIntegration {
     this.name = 'codex';
     this.displayName = 'OpenAI Codex CLI';
     this.bridgeFiles = ['AGENTS.md'];
-    this.generatedFiles = [];
-    this.platformDir = '';
+    this.generatedFiles = [
+      '.codex/skills/analyze/SKILL.md',
+      '.codex/skills/review/SKILL.md',
+      '.codex/skills/refactor/SKILL.md',
+      '.codex/skills/test/SKILL.md',
+      '.codex/skills/doc/SKILL.md',
+      '.codex/skills/spec/SKILL.md',
+      '.codex/skills/spec-component/SKILL.md',
+      '.codex/skills/ck/SKILL.md',
+      '.codex/skills/squad/SKILL.md',
+      '.codex/skills/squad-architect/SKILL.md',
+      '.codex/skills/squad-dev/SKILL.md',
+      '.codex/skills/squad-test/SKILL.md',
+      '.codex/skills/squad-review/SKILL.md',
+      '.codex/skills/squad-auto/SKILL.md',
+      '.codex/skills/squad-reset/SKILL.md',
+      '.codex/skills/squad-doc/SKILL.md',
+      '.codex/skills/squad-go/SKILL.md',
+      '.codex/skills/squad-spec/SKILL.md',
+      '.codex/skills/doc-arch/SKILL.md',
+      '.codex/skills/doc-feature/SKILL.md',
+      '.codex/skills/doc-component/SKILL.md',
+      '.codex/skills/agent-push-checklist/SKILL.md',
+      '.codex/skills/context-budget/SKILL.md',
+      '.codex/skills/standards-aware/SKILL.md',
+    ];
+    this.platformDir = '.codex';
   }
 
   async generateFiles() {
-    // Bridge file: AGENTS.md (auto-loaded by Codex CLI)
+    // Bridge file: AGENTS.md (auto-loaded by Codex)
     const bridgeContent = `# Project Standards (ContextKit)
 
 This project uses [ContextKit](https://github.com/nolrm/contextkit) for AI development standards.
 
-${this.getStandardsBlock()}
+## Project Standards
+
+Read these files before writing code:
+
+- \`.contextkit/standards/code-style.md\` — Coding conventions and style rules
+- \`.contextkit/standards/testing.md\` — Testing patterns and requirements
+- \`.contextkit/standards/architecture.md\` — Architecture decisions and patterns
+- \`.contextkit/standards/ai-guidelines.md\` — AI behavior and usage guidelines
+- \`.contextkit/standards/workflows.md\` — Development workflows and processes
+- \`.contextkit/standards/glossary.md\` — Project terminology and shortcuts
+
+## Product Context
+
+- \`.contextkit/product/mission-lite.md\` — Product mission (condensed)
+- \`.contextkit/product/decisions.md\` — Architecture Decision Records
+- \`.contextkit/product/roadmap.md\` — Development roadmap
 
 ## Corrections Log
 
@@ -25,20 +65,357 @@ ${this.getStandardsBlock()}
 
 ## Quick Reference
 
-Before writing code, check the relevant standards files above. Always follow the project's established patterns and conventions.`;
+Before writing code, always read the relevant standards files above. ContextKit skills are available in \`.codex/skills/\` — use them to run the spec and squad pipelines.`;
 
     await this.writeBridgeFile('AGENTS.md', bridgeContent);
+
+    // Skills
+    await this.writeGeneratedFile(
+      '.codex/skills/analyze/SKILL.md',
+      `---
+name: analyze
+description: Analyze project and generate customized standards
+---
+
+Read \`.contextkit/commands/dev/analyze.md\` and execute the analysis workflow for this project.
+
+Scan the codebase structure, detect frameworks and patterns, then generate customized standards files in \`.contextkit/standards/\`.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/review/SKILL.md',
+      `---
+name: review
+description: Review current changes for correctness and standards compliance
+---
+
+Read \`.contextkit/commands/dev/review.md\` and execute the review workflow.
+
+Review current changes for correctness, standards compliance, and potential issues.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/refactor/SKILL.md',
+      `---
+name: refactor
+description: Improve code structure without changing behavior
+---
+
+Read \`.contextkit/commands/dev/refactor.md\` and execute the refactoring workflow.
+
+Improve code structure without changing behavior, keeping tests green at every step.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/test/SKILL.md',
+      `---
+name: test
+description: Generate or run tests covering happy paths, edge cases, and errors
+---
+
+Read \`.contextkit/commands/dev/run-tests.md\` and execute the testing workflow.
+
+Generate or run tests for the specified code, covering happy paths, edge cases, and errors.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/doc/SKILL.md',
+      `---
+name: doc
+description: Add inline docs, README sections, and usage examples
+---
+
+Read \`.contextkit/commands/docs/add-documentation.md\` and execute the documentation workflow.
+
+Add inline docs, README sections, and usage examples for the specified code.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/spec-component/SKILL.md',
+      `---
+name: spec-component
+description: Write a component spec (MD-first) before coding begins
+---
+
+Read \`.contextkit/commands/dev/spec-component.md\` and execute the component spec workflow.
+
+Write a component spec (MD-first) before any code is created. Scaffold the spec file colocated with the component and wait for review before coding begins.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/spec/SKILL.md',
+      `---
+name: spec
+description: Turn a product overview into a reference spec — data model, API contracts, UX flows, and squad-ready stories
+---
+
+Read \`.contextkit/commands/spec/spec.md\` and execute the spec pipeline workflow.
+
+Single inline CTO pass per scope. Reads a product overview, identifies logical scopes, and writes a full SPEC.md per scope covering data model, API contracts, UX flows, stories, and copy-paste squad commands. Run once per scope — picks up where it left off automatically.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/ck/SKILL.md',
+      `---
+name: ck
+description: Check project setup, standards status, and integrations
+---
+
+Read \`.contextkit/commands/dev/health-check.md\` and execute the health check workflow.
+
+Check project setup, standards status, and integrations. Report what needs attention.
+`
+    );
+
+    // Squad skills
+    await this.writeGeneratedFile(
+      '.codex/skills/squad/SKILL.md',
+      `---
+name: squad
+description: Squad pipeline kickoff — create handoff file and write PO spec
+---
+
+Read \`.contextkit/commands/squad/squad.md\` and execute the squad kickoff workflow.
+
+Create the handoff file and write the PO spec for the given task. Pass the task description as input.
+
+After kickoff, run \`$squad-auto\` to run the full pipeline hands-free, or step through manually with \`$squad-architect\` → \`$squad-dev\` → \`$squad-test\` → \`$squad-review\` → \`$squad-doc\`.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/squad-architect/SKILL.md',
+      `---
+name: squad-architect
+description: Write technical implementation plan from PO spec (manual step 1/4)
+---
+
+Read \`.contextkit/commands/squad/squad-architect.md\` and execute the architect workflow.
+
+Read the PO spec from the handoff file and design the technical approach. Use \`$squad-auto\` to run all steps automatically.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/squad-dev/SKILL.md',
+      `---
+name: squad-dev
+description: Implement code changes following architect plan (manual step 2/4)
+---
+
+Read \`.contextkit/commands/squad/squad-dev.md\` and execute the dev workflow.
+
+Follow the architect's plan to implement the code changes. Use \`$squad-auto\` to run all steps automatically.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/squad-test/SKILL.md',
+      `---
+name: squad-test
+description: Write and run tests against acceptance criteria (manual step 3/4)
+---
+
+Read \`.contextkit/commands/squad/squad-test.md\` and execute the test workflow.
+
+Write and run tests against the PO's acceptance criteria. Use \`$squad-auto\` to run all steps automatically.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/squad-review/SKILL.md',
+      `---
+name: squad-review
+description: Review full handoff and write pass/needs-work verdict (manual step 4/4)
+---
+
+Read \`.contextkit/commands/squad/squad-review.md\` and execute the review workflow.
+
+Review the full handoff (spec, plan, implementation, tests) and write the final verdict. Use \`$squad-auto\` to run all steps automatically.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/squad-auto/SKILL.md',
+      `---
+name: squad-auto
+description: Auto-run full squad pipeline hands-free (architect → dev → test → review → doc)
+---
+
+Read \`.contextkit/commands/squad/squad-auto.md\` and execute the pipeline runner workflow.
+
+Run after \`$squad\` kickoff. Automatically runs architect → dev → test → review → doc for all tasks sequentially, hands-free.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/squad-reset/SKILL.md',
+      `---
+name: squad-reset
+description: Delete squad state to start fresh
+---
+
+Read \`.contextkit/commands/squad/squad-reset.md\` and execute the reset workflow.
+
+Delete the current squad state (\`.contextkit/squad/\`) so you can start fresh.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/squad-doc/SKILL.md',
+      `---
+name: squad-doc
+description: Document changes after review passes (manual step 5/5)
+---
+
+Read \`.contextkit/commands/squad/squad-doc.md\` and execute the doc workflow.
+
+After review passes, create or update companion .md files for every new/modified code file in this task.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/squad-go/SKILL.md',
+      `---
+name: squad-go
+description: Extract tasks from conversation and run the full pipeline immediately — no checkpoint pause
+---
+
+Read \`.contextkit/commands/squad/squad-go.md\` and execute the express pipeline workflow.
+
+Reads tasks from the current conversation, writes PO specs, and immediately runs architect → dev → test → review → doc. No checkpoint pause — single invocation, hands-free.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/squad-spec/SKILL.md',
+      `---
+name: squad-spec
+description: Implement every story in a spec scope — run /spec first to generate the scope
+---
+
+Read \`.contextkit/commands/squad/squad-spec.md\` and execute it.
+
+Processes one story per invocation. Pass the scope slug (e.g. \`01-identity-auth\`). Run \`$spec\` first to generate the scope, then \`$squad-spec [scope-slug]\` to implement it story by story.
+`
+    );
+
+    // Doc skills
+    await this.writeGeneratedFile(
+      '.codex/skills/doc-arch/SKILL.md',
+      `---
+name: doc-arch
+description: Generate or update architecture documentation (Level 1)
+---
+
+Read \`.contextkit/commands/docs/doc-arch.md\` and execute the architecture documentation workflow.
+
+Detect the project stack, then generate or update \`docs/architecture.md\` with system boundaries, key flows, and stack-appropriate artifacts.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/doc-feature/SKILL.md',
+      `---
+name: doc-feature
+description: Generate or update feature documentation (Level 2)
+---
+
+Read \`.contextkit/commands/docs/doc-feature.md\` and execute the feature documentation workflow.
+
+Detect the project stack, identify the target feature, then generate or update \`docs/features/<name>.md\`.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/doc-component/SKILL.md',
+      `---
+name: doc-component
+description: Generate or update component documentation (Level 3)
+---
+
+Read \`.contextkit/commands/docs/doc-component.md\` and execute the component documentation workflow.
+
+Detect the project stack, read the target file, then create or update a colocated \`<name>.md\`.
+`
+    );
+
+    // Agent skills
+    await this.writeGeneratedFile(
+      '.codex/skills/agent-push-checklist/SKILL.md',
+      `---
+name: agent-push-checklist
+description: Pre-push quality checklist for agents before git push
+---
+
+Read \`.contextkit/commands/agents/agent-push-checklist.md\` and execute the agent push checklist workflow.
+
+Run before pushing from an AI agent. Validates that changes are safe, tests pass, and the push is ready.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/context-budget/SKILL.md',
+      `---
+name: context-budget
+description: Check context consumption and advise on compact, summarise, or continue
+---
+
+Read \`.contextkit/commands/agents/context-budget.md\` and execute the context budget workflow.
+
+Check how much context has been consumed and advise on whether to compact, summarise, or continue.
+`
+    );
+
+    await this.writeGeneratedFile(
+      '.codex/skills/standards-aware/SKILL.md',
+      `---
+name: standards-aware
+description: Load and apply project standards before acting in an agentic context
+---
+
+Read \`.contextkit/commands/agents/standards-aware.md\` and execute the standards-aware workflow.
+
+Load and apply the project's ContextKit standards before taking action in an agentic context.
+`
+    );
   }
 
   showUsage() {
     console.log('');
     console.log(chalk.bold('  Codex CLI Usage:'));
-    console.log('    AGENTS.md is auto-loaded by Codex CLI');
-    console.log('    Just run: codex "create a button component"');
-    console.log('    Standards are automatically included via AGENTS.md');
+    console.log('    AGENTS.md is auto-loaded every session');
+    console.log('    Skills are installed in .codex/skills/');
     console.log('');
-    console.log(chalk.dim('  Files created:'));
-    console.log(chalk.dim('    AGENTS.md (bridge - auto-loaded)'));
+    console.log(chalk.dim('  Skills (use $skill-name in Codex):'));
+    console.log(chalk.dim('    $analyze       — Analyze project and generate standards'));
+    console.log(chalk.dim('    $review        — Review current changes'));
+    console.log(chalk.dim('    $refactor      — Refactor code structure'));
+    console.log(chalk.dim('    $test          — Generate or run tests'));
+    console.log(chalk.dim('    $doc           — Add documentation'));
+    console.log(chalk.dim('    $doc-arch      — Architecture docs (Level 1)'));
+    console.log(chalk.dim('    $doc-feature   — Feature docs (Level 2)'));
+    console.log(chalk.dim('    $doc-component — Component docs (Level 3)'));
+    console.log(chalk.dim(''));
+    console.log(chalk.dim('  Squad pipeline:'));
+    console.log(chalk.dim('    $squad "task"      — Kickoff: create handoff + PO spec'));
+    console.log(chalk.dim('    $squad-auto        — Auto-run full pipeline (recommended)'));
+    console.log(chalk.dim('    $squad-go          — Extract tasks from conversation and run'));
+    console.log(chalk.dim('    $squad-spec [scope] — Implement all stories in a spec scope'));
+    console.log(chalk.dim('    $squad-architect / $squad-dev / $squad-test / $squad-review'));
+    console.log(chalk.dim(''));
+    console.log(chalk.dim('  Spec pipeline:'));
+    console.log(chalk.dim('    $spec [scope]      — Run spec pipeline for next or named scope'));
+    console.log(chalk.dim(''));
+    console.log(chalk.dim('  Health check:'));
+    console.log(chalk.dim('    $ck                — Check project setup and standards status'));
   }
 }
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@nolrm/contextkit",
-  "version": "1.1.0",
+  "version": "1.1.1",
   "description": "ContextKit - Context Engineering + Agentic AI Pipelines. Scaffold structured standards for AI assistants and run autonomous multi-role pipelines (PO → Architect → Dev → Test → Review → Doc) grounded by that same context layer. Works with Cursor, Claude Code, Copilot, Codex, Gemini, Aider, and more.",
   "main": "lib/index.js",
   "bin": {