Add Claude local settings and update docs

Add .claude/settings.local.json to grant local Claude agent permissions for specific Bash commands. Remove the AI Operating Model section from AGENTS.md. Update CLAUDE.md to add npm run check:ecosystem and include it in the npm run check verification gate. Update README links to include the Code of Conduct and reorder docs links. Append [unreleased] and [1.0.0] references to CHANGELOG.md.

Author: bradpotts

.claude/settings.local.json

@@ -0,0 +1,14 @@
+{
+  "permissions": {
+    "allow": [
+      "Bash(npm run *)",
+      "Bash(npm test *)",
+      "Bash(npm install*)",
+      "Bash(npm ci)",
+      "Bash(git status)",
+      "Bash(git diff*)",
+      "Bash(git log*)",
+      "Bash(node *)"
+    ]
+  }
+}

AGENTS.md

@@ -29,28 +29,6 @@ public behavior or contract impact, and unresolved risks. Do not edit generated
 outputs directly. Do not update [CHANGELOG.md](CHANGELOG.md) unless the change
 is release-relevant.
 
-## AI Operating Model
-
-This is the central AI coordination document for the repository. Agent-specific
-files may add tool-local guidance, but they must not override the role
-boundaries below.
-
-This repository uses a five-agent AI operating model with defined,
-non-overlapping roles:
-
-| Agent              | Role                                                                   |
-| ------------------ | ---------------------------------------------------------------------- |
-| **Claude Code**    | Lead developer - primary implementation, architecture, tests           |
-| **OpenAI Codex**   | Documentation, releases, production stabilization, repo hygiene        |
-| **ChatGPT**        | Strategy, coordination, prompt design, and external review - support only |
-| **GitHub Copilot** | General development assistance (in-editor suggestions)                 |
-| **Google Jules**   | Automated maintenance - small fixes, dependency updates, micro-patches |
-
-Human final review, release decisions, tagging, publishing, and merge authority rest with Bradley Potts
-(brad.potts@coastdigitalgroup.com). Claude Code, Codex, and Copilot do not
-commit by default. Jules may commit and push only bounded automated
-maintenance when `JULES.md` explicitly allows it and all validation gates pass.
-
 ## Instruction Map
 
 | File                              | Audience                     | Purpose                                                            |

CHANGELOG.md

@@ -46,3 +46,6 @@ First public release.
 - `CLAUDE.md`, `CODEX.md`, Copilot instructions, Claude permissions, CodeRabbit review config, and the pull request template now align on AI-agent roles and release handoffs.
 - README AI guidance reference now points to `AGENTS.md` first.
 - Bumped `fs-extra` to `^11.3.5`, `@types/node` to `^25.7.0`, `typescript-eslint` to `^8.59.3`.
+
+[unreleased]: https://github.com/phcdevworks/spectre-init/compare/v1.0.0...HEAD
+[1.0.0]: https://github.com/phcdevworks/spectre-init/tree/v1.0.0

CLAUDE.md

@@ -51,7 +51,8 @@ npm run typecheck   # tsc --noEmit — catch type errors without building
 npm run lint        # ESLint flat config via jiti
 npm run build       # tsc → dist/
 npm run format      # Prettier write
-npm run check       # typecheck + lint + build (run before every commit)
+npm run check:ecosystem  # spectre-manifest schema + registration check
+npm run check       # typecheck + lint + build + check:ecosystem (run before every commit)
 ```
 
 There are no tests yet. `npm run check` is the full verification gate.

README.md

@@ -39,8 +39,8 @@
 
 CLI scaffolding for Spectre-ready applications. `spectre-init` creates a new TypeScript app from a bundled template, wires Spectre dependencies, and provides a predictable starter structure.
 
-[Contributing](CONTRIBUTING.md) | [Changelog](CHANGELOG.md) |
-[Roadmap](ROADMAP.md) | [Security Policy](SECURITY.md)
+[Contributing](CONTRIBUTING.md) | [Code of Conduct](CODE_OF_CONDUCT.md) |
+[Changelog](CHANGELOG.md) | [Roadmap](ROADMAP.md) | [Security Policy](SECURITY.md)
 
 ## When to use this package