@williambeto/ai-workflow 1.19.1 → 2.2.0

package/opencode/commands/specs/spec-to-tasks.md

@@ -1,26 +0,0 @@
-# OpenCode Command - Spec To Tasks
-
-## Purpose
-
-Convert an approved spec/change artifact into small, reviewable implementation tasks with validation and evidence expectations.
-
-## Constraints
-
-1. Do not implement.
-2. Keep tasks ordered and dependency-aware.
-3. Keep each task objective small and testable.
-4. Include validation criteria and expected evidence.
-
-## Context To Read
-
-- `templates/change-proposal.template.md`
-- `checklists/change-spec-readiness-checklist.md`
-- `AGENTS.md`
-
-## Expected Output
-
-- Ordered task list
-- Per task: objective, scope, out-of-scope, risk note
-- Validation criteria per task
-- Expected evidence per task
-- Recommended first task for implementation

package/opencode/commands/specs/spec-to-technical-plan.md

@@ -1,27 +0,0 @@
-# OpenCode Command - Spec To Technical Plan
-
-## Purpose
-
-Convert an approved specification into a technical plan safe for incremental implementation.
-
-## Context To Read
-
-- `docs/architecture-policy.md`
-- `docs/design-patterns-policy.md`
-- `AGENTS.md`
-- Approved specification
-
-## Constraints
-
-1. Do not implement.
-2. Keep architecture proportional.
-3. Keep design-pattern usage justified.
-4. Keep no-regression and validation explicit.
-
-## Expected Output
-
-- Technical approach summary
-- Risks and trade-offs
-- Validation strategy
-- PR-sequencing guidance
-- Blocking open questions

package/opencode/commands/start.md

@@ -1,45 +0,0 @@
-# Start workflow (default: orchestrator)
-
-Use this command when opening the project to pick the safest starting agent.
-
-## Default route
-
-Use `orchestrator` by default.
-
-## Dynamic fallback
-
-Switch from `orchestrator` only when one of these conditions is true:
-
-1. `discovery`
-   - Use when the request is vague, goals are unclear, or scope/constraints are missing.
-2. `prompt-engineer`
-   - Use when the request is mainly about prompts, commands, skills, or workflow-instruction quality.
-3. `planner`
-   - Use when user explicitly asks for planning-only output without execution/delegation.
-
-## Execution policy
-
-1. State chosen route in one line: `Selected agent: <name> — reason: <short reason>`.
-2. Enforce branch gate before any write operation:
-   - run `git status -sb`;
-   - if branch is `main`, create/switch to a scoped branch: `git switch -c <type>/<short-task-slug>`;
-   - continue write operations only off `main`.
-3. Keep no-regression and validation requirements explicit.
-4. If route confidence is low, ask at most two blocking questions.
-5. After route confirmation, proceed with that agent's normal workflow.
-6. Before delegating, classify task type and check if delegation is necessary (skip delegation for trivial one-file edits).
-7. Limit delegation to one owner + one reviewer unless task is explicitly cross-functional.
-
-## Expected output
-
-- Selected agent and reason
-- Scope assumptions
-- First safe next action
-
-## Auto-load routing
-
-After initial route confirmation:
-
-- if selected agent is `orchestrator`, delegate automatically by workflow gates and request confirmation only at required checkpoints;
-- if selected agent is `planner` and output includes a selected PR + handoff, auto-load `orchestrator` for gate-based step progression;
-- if selected agent is `discovery` or `prompt-engineer`, continue in that agent until output is ready for planner.

package/opencode/commands/token-economy.md

@@ -1,29 +0,0 @@
-# OpenCode Command - Token Economy
-
-## Purpose
-
-Apply compact output mode for current task without changing implementation behavior.
-
-## Context To Read
-
-- `AGENTS.md`
-- `.agents/skills/token-economy/SKILL.md`
-- current task prompt/handoff/issue
-
-## Instructions
-
-1. Select mode: `compact`, `terse`, `audit`, or `handoff`.
-2. Keep technical correctness, evidence, severity, validation commands, and acceptance criteria.
-3. Keep output short; remove filler and repeated explanations.
-4. Keep exact file paths, command names, and command results.
-5. Avoid broad repository scans unless explicitly requested.
-6. Do not alter workflow gates, branch safety, or validation rules.
-7. Do not commit unless explicitly requested.
-
-## Expected Output
-
-- Mode used
-- Task output in selected compact contract
-- Validation commands and results (if executed)
-- Risks
-- Next action

package/opencode/commands/validate.md

@@ -1,33 +0,0 @@
-# OpenCode Command - Validate
-
-## Purpose
-
-Validate the repository or current PR before commit or merge.
-
-## Context To Read
-
-- `package.json`
-- `.markdownlint.json`
-- `schemas/`
-
-## Instructions
-
-1. Inspect `package.json` for available validation scripts.
-2. Run available validation commands that are appropriate for the current change.
-3. Run Markdown validation when Markdown files changed.
-4. Check JSON parsing for `schemas/` when schemas exist or changed.
-5. Check Git status and the current diff.
-6. Summarize pass or fail with evidence.
-7. Identify blocking issues and non-blocking warnings.
-8. Avoid modifying files unless explicitly asked.
-9. Do not commit unless explicitly asked.
-
-## Expected Output
-
-- Validation status
-- Commands run
-- Results
-- Files changed
-- Blocking issues
-- Non-blocking warnings
-- Merge recommendation

package/opencode.jsonc

@@ -1,235 +0,0 @@
-{
-  "$schema": "https://opencode.ai/config.json",
-  "default_agent": "atlas",
-  "agent": {
-    "atlas": {
-      "mode": "primary",
-      "description": "Primary senior engineering partner and workflow orchestrator",
-      "prompt": "{file:./opencode/agents/atlas.md}"
-    },
-    "discovery": {
-      "mode": "primary",
-      "description": "Explore codebase, understand requirements, gather context",
-      "prompt": "{file:./opencode/agents/discovery.md}"
-    },
-    "planner": {
-      "mode": "primary",
-      "description": "Plan requirements, specs, technical plans, and PR breakdowns",
-      "prompt": "{file:./opencode/agents/planner.md}"
-    },
-    "spec-engineer": {
-      "mode": "primary",
-      "description": "Turn vague requests into lightweight spec-driven change artifacts",
-      "prompt": "{file:./opencode/agents/spec-engineer.md}"
-    },
-    "implementer": {
-      "mode": "primary",
-      "description": "Implement features, fixes, and code changes",
-      "prompt": "{file:./opencode/agents/implementer.md}"
-    },
-    "fixer": {
-      "mode": "primary",
-      "description": "Fix bugs, errors, and validation failures",
-      "prompt": "{file:./opencode/agents/fixer.md}"
-    },
-    "reviewer": {
-      "mode": "primary",
-      "description": "Review code, documentation, and implementation quality",
-      "prompt": "{file:./opencode/agents/reviewer.md}"
-    },
-    "validator": {
-      "mode": "primary",
-      "description": "Validate work against acceptance criteria and evidence",
-      "prompt": "{file:./opencode/agents/validator.md}"
-    },
-    "release-manager": {
-      "mode": "primary",
-      "description": "Prepare releases, commits, PR summaries, and deployments",
-      "prompt": "{file:./opencode/agents/release-manager.md}"
-    },
-    "wordpress-engineer": {
-      "mode": "primary",
-      "description": "Work on WordPress, WooCommerce, themes, and PHP 7 compatible code",
-      "prompt": "{file:./opencode/agents/wordpress-engineer.md}"
-    },
-    "prompt-engineer": {
-      "mode": "primary",
-      "description": "Design and refine prompts, skills, and workflow instructions",
-      "prompt": "{file:./opencode/agents/prompt-engineer.md}"
-    },
-    "orchestrator": {
-      "mode": "primary",
-      "description": "Route work across agents with gate-based handoff checks",
-      "prompt": "{file:./opencode/agents/orchestrator.md}"
-    },
-    "frontend-implementer": {
-      "mode": "subagent",
-      "description": "Implement frontend UI, components, state, routing, styling, accessibility",
-      "prompt": "{file:.agents/skills/frontend-implementer/SKILL.md}"
-    },
-    "backend-implementer": {
-      "mode": "subagent",
-      "description": "Implement backend APIs, services, persistence, authentication, validation",
-      "prompt": "{file:.agents/skills/backend-implementer/SKILL.md}"
-    },
-    "tester": {
-      "mode": "subagent",
-      "description": "Design and review tests, acceptance criteria, regression coverage",
-      "prompt": "{file:.agents/skills/tester/SKILL.md}"
-    },
-    "docs-writer": {
-      "mode": "subagent",
-      "description": "Write and improve documentation, README, guides, prompts, runbooks",
-      "prompt": "{file:.agents/skills/docs-writer/SKILL.md}"
-    },
-    "product-manager": {
-      "mode": "subagent",
-      "description": "Clarify requirements, define scope, write functional specifications",
-      "prompt": "{file:.agents/skills/product-manager/SKILL.md}"
-    },
-    "seo-audit": {
-      "mode": "subagent",
-      "description": "Audit SEO, AEO, and GEO signals with evidence-based findings",
-      "prompt": "{file:.agents/skills/seo-audit/SKILL.md}"
-    },
-    "tech-lead": {
-      "mode": "subagent",
-      "description": "Make technical decisions, review architecture, identify trade-offs",
-      "prompt": "{file:.agents/skills/tech-lead/SKILL.md}"
-    },
-    "deploy-engineer": {
-      "mode": "subagent",
-      "description": "Plan and review deployment, release, rollback, CI/CD readiness",
-      "prompt": "{file:.agents/skills/deploy-engineer/SKILL.md}"
-    },
-    "pr-orchestrator": {
-      "mode": "subagent",
-      "description": "Plan and validate small PRs, preserve scope, avoid regressions",
-      "prompt": "{file:.agents/skills/pr-orchestrator/SKILL.md}"
-    },
-    "build-and-validate": {
-      "mode": "subagent",
-      "description": "Run lint, typecheck, tests, build, and release readiness checks",
-      "prompt": "{file:.agents/skills/build-and-validate/SKILL.md}"
-    },
-    "napkin": {
-      "mode": "subagent",
-      "description": "Manage discoverable project memory for durable decisions",
-      "prompt": "{file:.agents/skills/napkin/SKILL.md}"
-    },
-    "minimal-context": {
-      "mode": "subagent",
-      "description": "Read and search repository with minimal token usage",
-      "prompt": "{file:.agents/skills/minimal-context/SKILL.md}"
-    },
-    "token-economy": {
-      "mode": "subagent",
-      "description": "Keep outputs compact while preserving technical accuracy",
-      "prompt": "{file:.agents/skills/token-economy/SKILL.md}"
-    },
-    "interface-design": {
-      "mode": "subagent",
-      "description": "Design product interfaces, dashboards, admin panels, tools",
-      "prompt": "{file:.agents/skills/interface-design/SKILL.md}"
-    },
-    "opencode-agent-design": {
-      "mode": "subagent",
-      "description": "Design OpenCode agents, commands, routing, role boundaries",
-      "prompt": "{file:.agents/skills/opencode-agent-design/SKILL.md}"
-    },
-    "stack-variant-creator": {
-      "mode": "subagent",
-      "description": "Create workflow variants for specific tech stacks",
-      "prompt": "{file:.agents/skills/stack-variant-creator/SKILL.md}"
-    },
-    "vue-nuxt": {
-      "mode": "subagent",
-      "description": "Structure Vue/Nuxt frontend — architecture, dashboards, CRUD, stores, composables, TypeScript contracts",
-      "prompt": "{file:.agents/skills/vue-nuxt/SKILL.md}"
-    },
-    "playwright-cli": {
-      "mode": "subagent",
-      "description": "Run and debug Playwright CLI workflows for browser tests",
-      "prompt": "{file:.agents/skills/playwright-cli/SKILL.md}"
-    }
-  },
-  "mcp": {
-    "linear": {
-      "type": "local",
-      "command": ["npx", "-y", "mcp-remote", "https://mcp.linear.app/mcp"],
-      "enabled": true
-    }
-  },
-  "command": {
-    "start": {
-      "description": "Select default startup route (orchestrator with fallback)",
-      "agent": "orchestrator",
-      "template": "{file:./opencode/commands/start.md}"
-    },
-    "ship": {
-      "description": "Run low-touch end-to-end workflow with required confirmations",
-      "agent": "orchestrator",
-      "template": "{file:./opencode/commands/ship.md}"
-    },
-    "plan": {
-      "description": "Plan from requirement to PR breakdown",
-      "agent": "planner",
-      "template": "{file:./opencode/commands/plan.md}"
-    },
-    "execute": {
-      "description": "Implement a selected PR",
-      "agent": "implementer",
-      "template": "{file:./opencode/commands/execute.md}"
-    },
-    "review": {
-      "description": "Review implementation or diff",
-      "agent": "reviewer",
-      "template": "{file:./opencode/commands/review.md}"
-    },
-    "validate": {
-      "description": "Validate work before commit or merge",
-      "agent": "validator",
-      "template": "{file:./opencode/commands/validate.md}"
-    },
-    "token-economy": {
-      "description": "Apply compact output mode for current task",
-      "agent": "orchestrator",
-      "template": "{file:./opencode/commands/token-economy.md}"
-    },
-    "autopilot": {
-      "description": "Run local-first task routing through existing workflow gates",
-      "agent": "orchestrator",
-      "template": "{file:./opencode/commands/autopilot.md}"
-    },
-    "roadmap-audit": {
-      "description": "Audit roadmap status and recommend smallest safe next PR",
-      "agent": "orchestrator",
-      "template": "{file:./opencode/commands/roadmap-audit.md}"
-    },
-    "orchestrate": {
-      "description": "Automatically route work across agents by gate status",
-      "agent": "orchestrator",
-      "template": "{file:./opencode/commands/orchestrate.md}"
-    },
-    "deploy": {
-      "description": "Prepare deployment readiness and rollback plan",
-      "agent": "deploy-engineer",
-      "template": "{file:./opencode/commands/deploy.md}"
-    },
-    "spec-create": {
-      "description": "Create a lightweight spec/change artifact from a vague request",
-      "agent": "spec-engineer",
-      "template": "{file:./opencode/commands/specs/create-spec-from-request.md}"
-    },
-    "spec-review": {
-      "description": "Review or improve spec readiness for implementation handoff",
-      "agent": "spec-engineer",
-      "template": "{file:./opencode/commands/specs/review-spec.md}"
-    },
-    "spec-tasks": {
-      "description": "Convert approved spec into small implementation tasks",
-      "agent": "spec-engineer",
-      "template": "{file:./opencode/commands/specs/spec-to-tasks.md}"
-    }
-  }
-}

package/packages/ai-workflow/README.md

@@ -1,82 +0,0 @@
-# @williambeto/ai-workflow
-
-`@williambeto/ai-workflow` is the npm CLI for installing AI Workflow Kit assets into a project.
-
-It prepares a repository for Codex and OpenCode workflows with local prompts, generated workflow assets, managed `opencode.jsonc` sections, and doctor checks.
-
-## Commands
-
-```bash
-npx @williambeto/ai-workflow init
-npx @williambeto/ai-workflow doctor
-npx @williambeto/ai-workflow codex:start
-npx @williambeto/ai-workflow codex:orchestrate
-```
-
-For existing projects, preview changes first:
-
-```bash
-npx @williambeto/ai-workflow init --dry-run
-```
-
-## Init flags
-
-```txt
---yes        non-interactive safe defaults
---force      replace managed files when conflicts exist
---dry-run    plan only; do not write files
---no-install skip dependency installation step
---no-overwrite skip conflicts even when --force is set
---profile=minimal|operational|full select onboarding profile (default: full)
-```
-
-## Notes
-
-- `init` uses `full` profile by default.
-- Setup is prepared for both Codex and OpenCode by default.
-- `init` creates or merges managed sections in `opencode.jsonc` for operational/full profiles.
-- `full` profile also installs local Codex prompt starter files (`start`, `orchestrate-next`, `plan`, `execute`, `review`, `validate`) under `.codex/prompts/`, plus OpenCode agents and skills under `opencode/agents/` and `.agents/skills/`.
-- conflict backups are stored under `.ai-workflow-backups/` with per-file retention.
-
-- `doctor` validates profile readiness (`.ai-workflow.json` and `opencode.jsonc` managed blocks for operational/full profiles).
-
-## Profiles
-
-| Profile | Use when | Generated assets |
-| --- | --- | --- |
-| `minimal` | Basic workflow docs and Codex placeholders only. | `.ai-workflow.json`, `README.workflow.md`, `.codex/prompts/README.md`, `opencode/README.md`. |
-| `operational` | Repeatable PR workflow with minimal local assets. | Minimal assets plus managed `opencode.jsonc` default agent and `start` command. |
-| `full` | Local starter files for the full role catalog. | Operational assets plus starter files under `.codex/prompts/`, `opencode/agents/`, and `.agents/skills/`. |
-
-`full` creates starter files for local customization; it does not mean every project should keep every agent or skill.
-
-## Codex wrapper commands
-
-Use wrapper commands to print generated prompt entrypoints for copy/paste into Codex:
-
-```txt
-codex:start
-codex:orchestrate
-codex:plan
-codex:execute
-codex:review
-codex:validate
-```
-
-If prompt files are missing, re-run:
-
-```bash
-npx @williambeto/ai-workflow init --yes --profile=full --force
-```
-
-## Preview status
-
-This CLI is in preview. It is validated for:
-
-- `init --dry-run`;
-- `init --yes`;
-- `doctor`;
-- managed `opencode.jsonc` merge with JSONC comments;
-- `npm run validate` in projects with a `validate` script.
-
-Before using it in a shared project, inspect generated files and commit only intentional changes.

package/packages/ai-workflow/src/cli.js

@@ -1,70 +0,0 @@
-import { runInit } from "./commands/init.js";
-import { runDoctor } from "./commands/doctor.js";
-import { runGuide } from "./commands/guide.js";
-import { getCodexCommands, runCodexCommand } from "./commands/codex.js";
-
-function printHelp() {
-  const codexCommands = getCodexCommands().map((cmd) => `  ai-workflow ${cmd}`).join("\n");
-
-  console.log(`ai-workflow
-
-Usage:
-  ai-workflow init [--yes] [--force] [--dry-run] [--no-install] [--no-overwrite] [--profile minimal|operational|full]
-  ai-workflow doctor
-  ai-workflow guide
-${codexCommands}
-
-Commands:
-  init      Install AI workflow defaults (Codex + OpenCode)
-  doctor    Verify local ai-workflow installation
-  guide     Interactive quick-start guide (detect assets, choose goal, see next steps)
-  codex:*   Print generated Codex starter prompt content
-`);
-}
-
-function parseFlags(args) {
-  const profileArg = args.find((arg) => arg.startsWith("--profile="));
-
-  return {
-    yes: args.includes("--yes"),
-    force: args.includes("--force"),
-    dryRun: args.includes("--dry-run"),
-    noInstall: args.includes("--no-install"),
-    noOverwrite: args.includes("--no-overwrite"),
-    profile: profileArg ? profileArg.replace("--profile=", "") : undefined
-  };
-}
-
-export async function runCli(args) {
-  const [command] = args;
-
-  if (!command || command === "--help" || command === "-h") {
-    printHelp();
-    return;
-  }
-
-  if (command === "init") {
-    await runInit({
-      cwd: process.cwd(),
-      ...parseFlags(args.slice(1))
-    });
-    return;
-  }
-
-  if (command === "doctor") {
-    await runDoctor({ cwd: process.cwd() });
-    return;
-  }
-
-  if (command === "guide") {
-    await runGuide({ cwd: process.cwd() });
-    return;
-  }
-
-  if (getCodexCommands().includes(command)) {
-    await runCodexCommand({ cwd: process.cwd(), command });
-    return;
-  }
-
-  throw new Error(`unknown command: ${command}`);
-}

package/packages/ai-workflow/src/commands/codex.js

@@ -1,37 +0,0 @@
-import fs from "node:fs/promises";
-import path from "node:path";
-
-const CODEX_PROMPT_MAP = {
-  "codex:start": ".codex/prompts/start-project.md",
-  "codex:orchestrate": ".codex/prompts/orchestrate-next.md",
-  "codex:plan": ".codex/prompts/plan-from-requirement.md",
-  "codex:execute": ".codex/prompts/execute-selected-pr.md",
-  "codex:review": ".codex/prompts/review-implementation.md",
-  "codex:validate": ".codex/prompts/validate-work.md"
-};
-
-export function getCodexCommands() {
-  return Object.keys(CODEX_PROMPT_MAP);
-}
-
-export async function runCodexCommand({ cwd, command }) {
-  const relativePath = CODEX_PROMPT_MAP[command];
-
-  if (!relativePath) {
-    throw new Error(`unknown codex command: ${command}`);
-  }
-
-  const absolutePath = path.join(cwd, relativePath);
-
-  let content;
-  try {
-    content = await fs.readFile(absolutePath, "utf8");
-  } catch {
-    throw new Error(
-      `${relativePath} not found. Run: npx ai-workflow init --yes --profile=full --force`
-    );
-  }
-
-  console.log(`# Source: ${relativePath}\n`);
-  console.log(content.trimEnd());
-}