enterprise-delivery 0.1.0

package/.claude-plugin/plugin.json

@@ -0,0 +1,21 @@
+{
+  "name": "enterprise-delivery",
+  "version": "0.1.0",
+  "description": "Repo-first agile delivery governance with validation gates.",
+  "skills": [
+    { "path": "skills/enterprise-kickoff" },
+    { "path": "skills/add-requirement" },
+    { "path": "skills/change-requirement" },
+    { "path": "skills/analyze-requirement" },
+    { "path": "skills/analyze-tech-stack" },
+    { "path": "skills/update-coding-rules" },
+    { "path": "skills/start-task" },
+    { "path": "skills/complete-task" },
+    { "path": "skills/story-intake" },
+    { "path": "skills/delivery-planning" },
+    { "path": "skills/delivery-lead" },
+    { "path": "skills/feature-gate" },
+    { "path": "skills/sprint-gate" },
+    { "path": "skills/release-gate" }
+  ]
+}

package/.codex-plugin/plugin.json

@@ -0,0 +1,36 @@
+{
+  "name": "enterprise-delivery",
+  "version": "0.1.0",
+  "description": "Repo-first agile delivery governance with story, change, sprint, release, and validation gates.",
+  "author": {
+    "name": "unknown"
+  },
+  "license": "unknown",
+  "keywords": [
+    "delivery",
+    "governance",
+    "requirements",
+    "traceability",
+    "validation",
+    "codex"
+  ],
+  "skills": "./skills/",
+  "interface": {
+    "displayName": "Enterprise Delivery",
+    "shortDescription": "Repo-first delivery governance for agentic software work",
+    "longDescription": "Use Enterprise Delivery to manage requirements, story and change records, risk, sprint and release traceability, and validation gates from repository artifacts.",
+    "developerName": "unknown",
+    "category": "Coding",
+    "capabilities": [
+      "Interactive",
+      "Read",
+      "Write"
+    ],
+    "defaultPrompt": [
+      "Initialize Enterprise Delivery for this repository.",
+      "Start delivery tracking for this feature."
+    ],
+    "brandColor": "#2563EB",
+    "screenshots": []
+  }
+}

package/AGENTS.md

@@ -0,0 +1,73 @@
+# Agent Workflow
+
+This repository uses Enterprise Delivery as the delivery source of truth. Follow this workflow for any delivery-impacting change, including features, bug fixes, validator behavior changes, documentation changes that affect users, release work, or project tracking updates.
+
+## Source Of Truth
+
+- Keep canonical delivery state in `docs/enterprise/`.
+- External ticket systems, pull requests, chats, and documents may be linked, but they are not the source of truth.
+- Do not invent stakeholders, approvals, roadmap commitments, sprint commitments, release readiness, or historical facts. Mark uncertain facts as `unknown` or `inferred`.
+
+## Required Chain
+
+Every delivery-impacting change must preserve this chain:
+
+```text
+Requirement -> STORY-* -> CHG-* -> tests/evidence -> sprint/release traceability
+```
+
+## Before Implementation
+
+1. Create or update a story under `docs/enterprise/stories/`.
+2. Create or update a linked change record under `docs/enterprise/changes/`.
+3. Add the change ID to story frontmatter `linked_changes`.
+4. Add the story ID to change frontmatter `related_stories`.
+5. Record requirement source, business value, acceptance criteria, docs impact, risk impact, and planned test evidence.
+
+Implementation should not start until the story and linked change exist for delivery-impacting work.
+
+## During Implementation
+
+- Keep the story and change record current when scope changes.
+- Update `docs/enterprise/risk-register.md` for new or changed delivery risks.
+- Update `docs/enterprise/decision-log.md` for decisions that affect architecture, delivery, release readiness, or governance.
+- Add command-based test evidence when verification is run.
+
+## Completion Gates
+
+Before claiming a story or feature is complete, run:
+
+```bash
+node scripts/enterprise-delivery-validate.mjs validate feature --story STORY-0001
+```
+
+Before claiming sprint state is ready, reviewed, or closed, run:
+
+```bash
+node scripts/enterprise-delivery-validate.mjs validate sprint --sprint 2026-W20
+```
+
+Before claiming release readiness, run:
+
+```bash
+node scripts/enterprise-delivery-validate.mjs validate release --release 2026.05.0
+```
+
+Before claiming repository delivery setup is valid, run:
+
+```bash
+node scripts/enterprise-delivery-validate.mjs validate repo
+```
+
+Treat validator exit code `1` as a delivery blocker. Fix the artifacts or evidence, then rerun the gate.
+
+## Required Verification
+
+For code changes in this repository, run:
+
+```bash
+npm test
+node scripts/enterprise-delivery-validate.mjs validate repo
+```
+
+Do not claim completion unless the relevant tests and enterprise gate pass, or clearly report what could not be run.

package/README.md

@@ -0,0 +1,127 @@
+# Enterprise Delivery
+
+Enterprise Delivery is a standalone plugin that adds repo-first agile delivery governance for agentic software work.
+
+It is designed to run alongside Superpowers. Superpowers handles engineering workflow. Enterprise Delivery handles requirements, backlog, sprint and release tracking, change tracking, traceability, risk and status reporting, and validation gates.
+
+## Principles
+
+- Repo-first source of truth.
+- External tools are links, not canonical state.
+- Every story has a linked `CHG-*` record from creation.
+- Validation gates require evidence before completion claims.
+- Brownfield onboarding labels facts as confirmed, inferred, or unknown.
+- No third-party runtime dependencies.
+
+## Target Project Artifacts
+
+Enterprise Delivery stores project artifacts in:
+
+```text
+docs/enterprise/
+  project-charter.md
+  roadmap.md
+  architecture-overview.md
+  onboarding-audit.md  # brownfield-only audit artifact
+  test-strategy.md
+  stakeholder-register.md
+  risk-register.md
+  decision-log.md
+  product-backlog.md
+  change-log.md
+  project-vision.md
+  current-state.md
+  task-graph.json
+  changes/
+  stories/
+  sprints/
+  releases/
+```
+
+## AI-Native State Layer
+
+For spec-driven AI-native projects, Enterprise Delivery also keeps project memory under `docs/enterprise/`:
+
+- `project-vision.md`: product goals, non-goals, architecture principles, forbidden patterns, coding philosophy, quality bar, and AI collaboration rules.
+- `current-state.md`: active stories and changes, failing tests, TODOs, coverage, unresolved bugs, recent commits, risks, decisions, and next verification commands.
+- `task-graph.json`: machine-readable task dependencies for humans and agents.
+
+These files are part of the same source of truth as stories, changes, risks, decisions, sprints, and releases. Do not create root-level project state files that drift from `docs/enterprise/`.
+
+## npm and Codex Install
+
+Install the package into your local Codex plugin marketplace:
+
+```bash
+npx enterprise-delivery install codex
+```
+
+The command copies the Enterprise Delivery plugin payload into:
+
+```text
+~/.agents/plugins/plugins/enterprise-delivery/
+```
+
+It also creates or updates:
+
+```text
+~/.agents/plugins/marketplace.json
+```
+
+The command is safe to rerun. It replaces the managed `enterprise-delivery` plugin copy and preserves unrelated marketplace entries.
+
+Claude support is reserved for a later adapter:
+
+```bash
+npx enterprise-delivery install claude
+```
+
+For now this command only reports that the Claude adapter is planned but not implemented.
+
+## Validation
+
+Initialize delivery artifacts in a target project repository:
+
+```bash
+node /path/to/enterprise-delivery/scripts/enterprise-delivery-validate.mjs init
+```
+
+The init command creates missing files under `docs/enterprise/` from templates and preserves existing files.
+
+Run validation from a target project repository:
+
+```bash
+node /path/to/enterprise-delivery/scripts/enterprise-delivery-validate.mjs validate repo
+node /path/to/enterprise-delivery/scripts/enterprise-delivery-validate.mjs validate feature --story STORY-0001
+node /path/to/enterprise-delivery/scripts/enterprise-delivery-validate.mjs validate sprint --sprint 2026-W20
+node /path/to/enterprise-delivery/scripts/enterprise-delivery-validate.mjs validate release --release 2026.05.0
+```
+
+Use `--root PATH` to initialize or validate a different repository and `--json` for machine-readable output.
+
+Exit codes:
+
+- `0`: validation passed
+- `1`: validation failed because artifacts are missing or invalid
+- `2`: command usage was invalid
+
+## Skills
+
+- `enterprise-kickoff`: initialize or audit `docs/enterprise/`.
+- `add-requirement`: create a linked `STORY-*` and `CHG-*` from a new request.
+- `change-requirement`: revise existing requirement artifacts and trace changed scope.
+- `analyze-requirement`: clarify scope, risks, tasks, and planned evidence before coding.
+- `analyze-tech-stack`: document stack facts, tradeoffs, decisions, and risks.
+- `update-coding-rules`: update agent rules, project principles, quality gates, and workflow instructions.
+- `start-task`: begin implementation from a story/change/task graph entry.
+- `complete-task`: run verification, record evidence, and prepare feature completion.
+- `story-intake`: create or update story and mandatory change records.
+- `delivery-planning`: map acceptance criteria to tasks, tests, docs, and risks.
+- `delivery-lead`: guide agents through project start, requirement intake, requirement analysis, tech-stack decisions, coding rules, task start, and completion.
+- `feature-gate`: validate story completion evidence.
+- `sprint-gate`: validate sprint state and reporting.
+- `release-gate`: validate release traceability and readiness.
+
+## Brownfield Rule
+
+Do not fabricate enterprise history. Facts discovered from files may be marked inferred. Stakeholders, roadmap, sprint commitments, release state, and approvals require explicit human confirmation before they are marked confirmed.

package/package.json

@@ -0,0 +1,28 @@
+{
+  "name": "enterprise-delivery",
+  "version": "0.1.0",
+  "type": "module",
+  "bin": {
+    "enterprise-delivery": "scripts/enterprise-delivery-validate.mjs"
+  },
+  "files": [
+    ".claude-plugin/",
+    ".codex-plugin/",
+    "AGENTS.md",
+    "README.md",
+    "package.json",
+    "scripts/",
+    "skills/",
+    "src/",
+    "templates/"
+  ],
+  "scripts": {
+    "test": "bash tests/run-tests.sh",
+    "init": "node scripts/enterprise-delivery-validate.mjs init",
+    "validate": "node scripts/enterprise-delivery-validate.mjs",
+    "gate:repo": "node scripts/enterprise-delivery-validate.mjs validate repo",
+    "gate:feature": "node scripts/enterprise-delivery-validate.mjs validate feature",
+    "gate:sprint": "node scripts/enterprise-delivery-validate.mjs validate sprint",
+    "gate:release": "node scripts/enterprise-delivery-validate.mjs validate release"
+  }
+}

package/scripts/enterprise-delivery-validate.mjs

@@ -0,0 +1,11 @@
+#!/usr/bin/env node
+import { runCli } from '../src/cli.mjs';
+
+runCli(process.argv.slice(2), process.cwd())
+  .then((exitCode) => {
+    process.exit(exitCode);
+  })
+  .catch((error) => {
+    console.error(error && error.stack ? error.stack : String(error));
+    process.exit(2);
+  });

package/skills/add-requirement/SKILL.md

@@ -0,0 +1,52 @@
+---
+name: add-requirement
+description: Use when the user says add requirement, create requirement, capture requirement, new feature request, new bug request, or asks to turn a request into repo-tracked delivery artifacts.
+---
+
+# Add Requirement
+
+Create the delivery source of truth before implementation.
+
+## Read
+
+- `AGENTS.md`
+- `docs/enterprise/project-vision.md`
+- `docs/enterprise/current-state.md`
+- `docs/enterprise/product-backlog.md`
+- existing `docs/enterprise/stories/STORY-*.md`
+- existing `docs/enterprise/changes/CHG-*.md`
+
+## Create Or Update
+
+- `docs/enterprise/stories/STORY-<next>-<slug>.md`
+- `docs/enterprise/changes/CHG-<next>-<slug>.md`
+- `docs/enterprise/product-backlog.md`
+- `docs/enterprise/current-state.md`
+- `docs/enterprise/task-graph.json` only if initial tasks are already clear
+
+## Required Story Content
+
+- Requirement source: quote or summarize the user request.
+- Business value: why it matters.
+- Acceptance criteria: reviewable behavior.
+- Non-functional criteria: performance, security, accessibility, reliability, or no new criteria.
+- Dependencies: known dependencies or no known dependencies.
+- Risk impact: risk-register reference or no new risks.
+- Docs impact: docs changed or `No impact: <reason>`.
+- Test evidence: planned commands.
+- Implementation links: pending.
+
+## Required Change Content
+
+- Reason
+- Impact
+- Approval
+- Evidence plan
+- Rollout notes
+
+## Rules
+
+- Add the change ID to story frontmatter `linked_changes`.
+- Add the story ID to change frontmatter `related_stories`.
+- Keep story status `draft` until the source and acceptance criteria are explicit.
+- Do not implement code in this skill unless the user explicitly asks to continue.

package/skills/analyze-requirement/SKILL.md

@@ -0,0 +1,48 @@
+---
+name: analyze-requirement
+description: Use when the user asks to analyze a requirement, refine a requirement, check feasibility, split scope, identify risks, map acceptance criteria to tasks, or prepare a requirement for implementation.
+---
+
+# Analyze Requirement
+
+Turn a requirement into implementation-ready delivery context.
+
+## Read
+
+- `AGENTS.md`
+- `docs/enterprise/project-vision.md`
+- `docs/enterprise/current-state.md`
+- target `docs/enterprise/stories/STORY-*.md`
+- linked `docs/enterprise/changes/CHG-*.md`
+- `docs/enterprise/architecture-overview.md`
+- `docs/enterprise/test-strategy.md`
+- `docs/enterprise/risk-register.md`
+- `docs/enterprise/decision-log.md`
+- `docs/enterprise/task-graph.json`
+
+## Do
+
+- Identify unclear terms and mark them as questions, assumptions, `unknown`, or `inferred`.
+- Check whether acceptance criteria are testable.
+- Split independent outcomes into separate stories.
+- Map each acceptance criterion to tasks and verification commands.
+- Identify docs, risk, decision, migration, and rollout impacts.
+
+## Update
+
+- target `STORY-*`
+- linked `CHG-*`
+- `docs/enterprise/task-graph.json`
+- `docs/enterprise/risk-register.md` if risks are added or changed
+- `docs/enterprise/decision-log.md` if decisions are made
+- `docs/enterprise/current-state.md`
+
+## Output
+
+Give the user:
+
+- clarified requirement summary
+- open questions
+- recommended scope
+- task graph changes
+- planned evidence commands

package/skills/analyze-tech-stack/SKILL.md

@@ -0,0 +1,36 @@
+---
+name: analyze-tech-stack
+description: Use when the user asks to choose, review, compare, or document the project tech stack, architecture choices, frameworks, databases, hosting, testing tools, or provider integrations.
+---
+
+# Analyze Tech Stack
+
+Document stack facts and decisions without inventing commitments.
+
+## Read
+
+- `AGENTS.md`
+- `docs/enterprise/project-charter.md`
+- `docs/enterprise/project-vision.md`
+- `docs/enterprise/architecture-overview.md`
+- `docs/enterprise/test-strategy.md`
+- `docs/enterprise/risk-register.md`
+- `docs/enterprise/decision-log.md`
+- package/build files such as `package.json`, `pyproject.toml`, `Cargo.toml`, `go.mod`
+- existing source and test layout when relevant
+
+## Update
+
+- `docs/enterprise/architecture-overview.md`
+- `docs/enterprise/test-strategy.md`
+- `docs/enterprise/risk-register.md`
+- `docs/enterprise/decision-log.md`
+- active `STORY-*` and `CHG-*` if this analysis belongs to a delivery change
+
+## Rules
+
+- Mark implemented repo facts as `confirmed` or `inferred` depending on evidence.
+- Mark undecided choices as `unknown`.
+- Record durable choices in `decision-log.md`.
+- Record tradeoffs and risks in `risk-register.md`.
+- Do not add dependencies or change code unless the user explicitly asks for implementation.

package/skills/change-requirement/SKILL.md

@@ -0,0 +1,37 @@
+---
+name: change-requirement
+description: Use when the user says change requirement, update requirement, revise story, modify acceptance criteria, adjust scope, or correct an existing STORY-* or CHG-* requirement.
+---
+
+# Change Requirement
+
+Revise existing delivery artifacts without losing history or traceability.
+
+## Read
+
+- `AGENTS.md`
+- `docs/enterprise/project-vision.md`
+- `docs/enterprise/current-state.md`
+- target `docs/enterprise/stories/STORY-*.md`
+- linked `docs/enterprise/changes/CHG-*.md`
+- `docs/enterprise/task-graph.json`
+- `docs/enterprise/risk-register.md`
+- `docs/enterprise/decision-log.md`
+
+## Update
+
+- target `STORY-*`
+- linked `CHG-*`
+- `docs/enterprise/product-backlog.md` if priority/status changes
+- `docs/enterprise/task-graph.json` if tasks or dependencies change
+- `docs/enterprise/risk-register.md` if risk changes
+- `docs/enterprise/decision-log.md` if architecture, governance, or delivery decisions change
+- `docs/enterprise/current-state.md`
+
+## Rules
+
+- Preserve the original requirement source in change history when useful.
+- Add a `Change History` entry describing what changed and why.
+- Update planned evidence when acceptance criteria change.
+- If scope becomes too broad, split into a new `STORY-*` and linked `CHG-*`.
+- Do not mark changed work complete until evidence is rerun.

package/skills/complete-task/SKILL.md

@@ -0,0 +1,46 @@
+---
+name: complete-task
+description: Use when the user asks to complete a task, finish a feature, close a story, verify work, record evidence, or prepare delivery completion for a STORY-* or CHG-*.
+---
+
+# Complete Task
+
+Record evidence before claiming completion.
+
+## Read
+
+- `AGENTS.md`
+- target `docs/enterprise/stories/STORY-*.md`
+- linked `docs/enterprise/changes/CHG-*.md`
+- `docs/enterprise/task-graph.json`
+- `docs/enterprise/current-state.md`
+- `docs/enterprise/test-strategy.md`
+
+## Run
+
+Use the story's planned evidence commands. For code changes in this repo, run at minimum:
+
+```bash
+npm test
+node scripts/enterprise-delivery-validate.mjs validate repo
+node scripts/enterprise-delivery-validate.mjs validate feature --story STORY-0001
+```
+
+Replace `STORY-0001` with the target story ID.
+
+## Update
+
+- story `Test Evidence` with actual commands and results
+- story `Implementation Links`
+- change `Evidence` with actual commands and results
+- change status to `implemented`, `verified`, or `released`
+- task status and evidence in `docs/enterprise/task-graph.json`
+- `docs/enterprise/current-state.md`
+- `docs/enterprise/risk-register.md` if risks changed
+- `docs/enterprise/decision-log.md` if decisions changed
+
+## Rules
+
+- Do not claim completion if validation exits `1`.
+- If tests fail, record the failure in `current-state.md` and keep the story out of `done`.
+- Only run sprint or release gates when the user asks for sprint or release readiness.