{
  "markdown": "# Team AI Directives Starter Kit\n\nA forkable foundation for version-controlled AI agent behavior.\n\n**Built on the [Twelve-Factor Agentic SDLC](https://github.com/tikalk/agentic-sdlc-12-factors)** — this repository implements Factor XI: Directives as Code, treating all AI instructions as version-controlled assets.\n\n**How the pieces fit:**\n\n- **[12-Factor Agentic SDLC](https://github.com/tikalk/agentic-sdlc-12-factors)** — the methodology (strategic mindset, structured planning, directives as code, traceability)\n- **This repo** — the version-controlled team knowledge base (constitution, personas, rules, skills, CDRs)\n- **[agentic-sdlc-spec-kit](https://github.com/tikalk/agentic-sdlc-spec-kit)** — the Spec-Driven Development toolkit; consumes this knowledge base via the bundled `team-ai-directives` extension\n- **[adlc-team-skills](https://github.com/tikalk/adlc-team-skills)** — agent skills that implement the methodology; consume this knowledge base via `team-*` and `levelup-*` skills\n\n**Quick Start:** See [GETTING_STARTED.md](GETTING_STARTED.md) for a 5-minute setup guide.\n\n## Installation\n\nThis knowledge base is consumed in two ways at runtime — pick the one that matches your project setup. Fork and clone for authoring and customization.\n\n### Option 1: Spec Kit Projects\n\nFor projects managed with the [Agentic SDLC Spec Kit](https://github.com/tikalk/agentic-sdlc-spec-kit), install via the `specify` CLI using the `--team-ai-directives` flag:\n\n```bash\n# Initialize project with team-ai-directives\nspecify init <project> --team-ai-directives https://github.com/your-org/team-ai-directives.git\n```\n\nThe Specify CLI installs the bundled `team-ai-directives` extension (governance commands and skills) and copies this repository's `default` domain skills into the agent's skills directory. Context modules are referenced through the `agent-context` extension.\n\n```bash\n# Or from a specific release tag\nspecify init <project> --team-ai-directives https://github.com/your-org/team-ai-directives/archive/refs/tags/v1.3.0.zip\n```\n\n### Option 2: Any Skills-Capable Agent\n\nFor agents that support the [Agent Skills standard](https://agentskills.io) (Claude Code, Codex, OpenCode, Cursor, Gemini, and others), install the governance and architecture skills from [adlc-team-skills](https://github.com/tikalk/adlc-team-skills):\n\n```bash\nnpx skills add tikalk/adlc-team-skills\n```\n\nThen invoke the `team-setup` skill in your project to clone, point at, or scaffold this knowledge base. The skills locate it via `.adlc/init-options.json` or the `ADLC_TEAM_AI_DIRECTIVES` environment variable.\n\n### Option 3: Authoring (Fork and Clone)\n\nFork this repository and clone it locally to customize the knowledge base itself:\n\n```bash\ngit clone https://github.com/your-org/team-ai-directives.git\ncd team-ai-directives\n```\n\nThen reference it in your project initialization or configuration.\n\n## Who This Is For\n\n- Engineering teams wanting consistent AI agent behavior\n- Platform teams building developer self-service with AI\n- Consultancies creating reusable AI patterns across clients\n- Organizations adopting the Agentic SDLC methodology\n\n## What's Included\n\nThis repository provides the building blocks for teaching AI agents how your team works:\n\n- **Constitution** - Core principles that govern all AI behavior\n- **Personas** - Role-specific guidance (DevOps, Java, Python, Data, Platform)\n- **Rules** - Domain-specific patterns (security, testing, style guides)\n- **Skills** - Self-contained capabilities with trigger-based activation\n\n## Repository Layout\n\n```text\nteam-ai-directives/\n├── AGENTS.md                  # Instructions for AI agents on how to use this repo\n├── README.md                  # This file (for humans)\n├── GETTING_STARTED.md         # Quick start guide\n├── CONTRIBUTING.md            # Contribution guidelines\n├── CHANGELOG.md               # Version history\n├── CDR.md                     # Context Directive Records (approved contributions)\n├── .mcp.json                  # MCP server configuration\n├── .skills.json               # Skills registry and policy\n├── context_modules/           # The knowledge library\n│   ├── constitution.md        # Core principles\n│   ├── personas/              # Role-specific guidance\n│   ├── rules/                 # Domain-specific patterns\n│   │   ├── style_guides/\n│   │   ├── security/\n│   │   └── testing/\n│   └── examples/              # Code examples and prompt templates\n│       ├── testing/\n│       └── prompts/\n└── skills/                    # Self-contained agent capabilities\n    └── {skill-name}/\n        ├── SKILL.md           # Main instructions with YAML frontmatter\n        ├── references/        # Supporting content\n        └── scripts/           # Automation (optional)\n```\n\n## File Format\n\nAll directives (rules, personas, examples, skills) published via LevelUp include **YAML frontmatter** for memory management:\n\n```yaml\n---\nid: rule-python-error-handling\ncdr_ref: CDR-2026-001\ncreated: 2026-04-15\nmodified: 2026-05-18\nverified: 2026-05-18\nage_days: 33\nevidence:\n  - commit: abc123\n    file: src/error_handler.py\n---\n```\n\n### Frontmatter Fields\n\n| Field | Description | Example |\n|-------|-------------|---------|\n| `id` | Unique identifier | `rule-python-error-handling` |\n| `cdr_ref` | Source CDR reference | `CDR-2026-001` |\n| `created` | Original publication date | `2026-04-15` |\n| `modified` | Last edit date | `2026-05-18` |\n| `verified` | Last verification date | `2026-05-18` |\n| `age_days` | Days since creation | `33` |\n| `evidence` | List of supporting commits/files | YAML list |\n\n### Freshness Warning\n\nPublished directives include a verification banner:\n\n```markdown\n> ⚠️ **Memory Verification**\n> This directive is 33 days old. Before applying:\n> - [ ] Pattern still exists in current codebase\n> - [ ] Rule is actively followed by team\n> - [ ] No conflicting rules introduced\n```\n\n### Verification Workflow\n\n1. Scan the knowledge base — `/levelup.validate` in spec-kit projects, or `team-repair --freshness` / `team-repair --conflicts` in skills-based projects\n2. Valid directives get their `verified` timestamp updated\n3. Stale directives (>30 days) are flagged for review\n4. Update or deprecate stale directives as needed\n\nSee [CONTRIBUTING.md](CONTRIBUTING.md) for the full verification workflow.\n\n## How It Works\n\n1. **AI agents read `AGENTS.md`** for instructions on using this repo\n2. **Constitution** provides foundational principles\n3. **Personas** provide role-specific guidance with rule references\n4. **Skills** are triggered by user requests (matched via `.skills.json`)\n5. **Rules** are accessed through personas, not directly from skills\n\n### Two Delivery Mechanisms\n\nThe same knowledge base is consumed through two complementary mechanisms:\n\n- **Spec Kit extension** — governance commands (canonical names `adlc.team-ai-directives.*`, invoked via short aliases like `team.discover`). Hooks auto-run `team.discover` before `specify` and `plan`. The knowledge base path is stored in `.specify/init-options.json`.\n- **Agent Skills ([adlc-team-skills](https://github.com/tikalk/adlc-team-skills))** — model-invoked skills following the [Agent Skills standard](https://agentskills.io). `team-boot` auto-loads the constitution at session start; `team-discover` auto-finds relevant context per task. The knowledge base path is resolved from `.adlc/init-options.json` or the `ADLC_TEAM_AI_DIRECTIVES` environment variable.\n\nBoth mechanisms read the same files — `AGENTS.md`, `CDR.md`, `.skills.json`, and `context_modules/` — and can coexist in one project.\n\n---\n\n## Personas\n\nPersonas define the role, expertise, preferences, and rule references that shape how an AI agent behaves for a given engineering context. Loading a persona tells the agent _who_ it is for this session — its values, collaboration style, and which domain-specific rules to apply.\n\n### How Personas Work\n\nWhen an agent begins a session, it loads context in this order:\n\n1. **Constitution** (`context_modules/constitution.md`) — non-negotiable team principles applied to every interaction.\n2. **Persona** (`context_modules/personas/*.md`) — role-specific defaults, rules, and collaboration style.\n3. **Skills** (`skills/*/SKILL.md`) — on-demand capabilities triggered by the user's request.\n\nA persona sits between the universal constitution and the task-specific skill. It tells the agent:\n- What domain knowledge to prioritize\n- Which rule files are relevant to its role\n- How to collaborate (communication style, review preferences, workflow assumptions)\n- Agent-specific guidance (e.g., always propose infra changes as code)\n\nPersonas are **passive by default** — they don't activate automatically unless your tooling or prompt instructs the agent to load one. You attach a persona to an agent through your IDE settings, a system prompt, or a prompt prefix.\n\n### Persona Folder Structure\n\nAll personas live in `context_modules/personas/`:\n\n```\ncontext_modules/personas/\n├── cloud_native_platform_architect.md\n├── data_analyst.md\n├── devops_engineer.md\n├── senior_java_developer.md\n└── senior_python_developer.md\n```\n\nEach persona is a single Markdown file. There is no sub-folder nesting — one file per role.\n\n### Persona File Anatomy\n\nA well-formed persona file contains some or all of the following sections:\n\n#### `# Persona: <Name>` (required)\n\nThe heading names the persona. Agents use this as the persona identifier.\n\n```markdown\n# Persona: DevOps Engineer\n```\n\n#### `## Summary` (required)\n\nDescribes the persona's **motivation**, **pain points**, and **success criteria**. This is the first context an agent reads to understand its role.\n\n```markdown\n## Summary\n- **Motivation**: Enable reliable, scalable, and secure software delivery through automation, IaC, and observability.\n- **Pain Points**: Manual deployments, configuration drift, lack of visibility.\n- **Success Criteria**: Fully automated CI/CD pipelines, declarative infrastructure, secure secret management.\n```\n\n#### `## Rule References` (recommended)\n\nLinks the persona to domain-specific rule files using the `@rule:<path>` syntax. Agents resolve these paths relative to `context_modules/rules/`.\n\n```markdown\n## Rule References\n- CI/CD Pipelines: @rule:devops/github_actions.md\n- Secrets Management: @rule:devops/external_secrets_operator.md, @rule:devops/secrets_management_dry.md\n- Testing: @rule:testing/python/pytest_patterns.md\n```\n\n#### `## Collaboration Preferences` (recommended)\n\nDescribes how the persona prefers to work: communication style, review expectations, workflow assumptions, and advocacy positions.\n\n```markdown\n## Collaboration Preferences\n- Prefers infrastructure changes reviewed through pull requests with clear descriptions\n- Values declarative configurations over imperative scripts\n- Expects \"everything as code\" — infrastructure, configs, and pipelines in version control\n```\n\n#### `## Tool Context` (optional)\n\nLists the tooling ecosystem this persona operates in. Helps agents make appropriate technology choices without asking the user every time.\n\n```markdown\n## Tool Context\n- CI: GitHub Actions, GitLab CI\n- CD/GitOps: ArgoCD, Flux\n- IaC: Terraform, Crossplane\n- Secrets: HashiCorp Vault, AWS Secrets Manager\n```\n\n#### `## Guidance for Agents` (optional)\n\nExplicit behavioral instructions for agents taking actions on behalf of this persona.\n\n```markdown\n## Guidance for Agents\n- Always propose infrastructure changes as code, never manual operations\n- When working with secrets, always use secret management services — never hardcode or commit secrets\n- Always consider disaster recovery, backup strategies, and rollback procedures\n```\n\n### Built-In Personas\n\n| File | Persona | Primary Domain |\n|---|---|---|\n| `senior_python_developer.md` | Senior Python Developer | Python, PEP 8, testing, CI/CD |\n| `senior_java_developer.md` | Senior Java Developer | Java, Spring Boot, JUnit 5, Google Style |\n| `devops_engineer.md` | DevOps Engineer | CI/CD, Helm, IaC, secrets management, GitOps |\n| `cloud_native_platform_architect.md` | Cloud-Native Platform Architect | Kubernetes, Crossplane, ArgoCD, platform engineering |\n| `data_analyst.md` | Data Analyst | SQL, dashboards, reproducible reporting, large datasets |\n\nEach built-in persona is self-contained and production-ready. Fork and adjust them to match your team's specific tooling and standards.\n\n### Creating a Custom Persona\n\n1. Create a new file in `context_modules/personas/`:\n\n   ```bash\n   touch context_modules/personas/my_role.md\n   ```\n\n2. Add the following template and fill it in:\n\n   ```markdown\n   # Persona: My Role\n\n   ## Summary\n   - **Motivation**: [What drives this role]\n   - **Pain Points**: [What slows them down]\n   - **Success Criteria**: [What good looks like]\n\n   ## Rule References\n   - [Domain]: @rule:[domain]/[rule-file].md\n\n   ## Collaboration Preferences\n   - [How this persona prefers to work]\n\n   ## Tool Context\n   - [Tools and platforms this persona uses]\n\n   ## Guidance for Agents\n   - [Behavioral instructions for autonomous actions]\n   ```\n\n3. Reference any applicable rule files from `context_modules/rules/`. Browse the available rules:\n\n   ```\n   context_modules/rules/\n   ├── architecture/\n   ├── data/\n   ├── devops/\n   ├── orchestration/\n   ├── security/\n   ├── style-guides/\n   │   ├── java/\n   │   └── python/\n   └── testing/\n       ├── java/\n       └── python/\n   ```\n\n4. Attach the persona in your agent configuration.\n\n### Using Personas\n\n#### Option A: IDE Custom Instructions (GitHub Copilot)\n\nIn VS Code with GitHub Copilot, add the persona content directly to a `.github/copilot-instructions.md` file or reference it in your workspace settings:\n\n```json\n// .vscode/settings.json\n{\n  \"github.copilot.chat.codeGeneration.instructions\": [\n    { \"file\": \"context_modules/personas/devops_engineer.md\" }\n  ]\n}\n```\n\n#### Option B: System Prompt Prefix\n\nPrepend the persona content to your agent's system prompt:\n\n```\n[Load persona: context_modules/personas/devops_engineer.md]\n\nUser request: ...\n```\n\n#### Option C: Prompt Reference at Session Start\n\nTell the agent explicitly which persona to adopt at the start of a conversation:\n\n```\nYou are acting as the DevOps Engineer persona defined in\ncontext_modules/personas/devops_engineer.md. Load that file\nand apply it to all responses in this session.\n```\n\n### Personas vs. Skills\n\n| | Persona | Skill |\n|---|---|---|\n| **Purpose** | Defines _who_ the agent is | Defines _what_ the agent can do |\n| **Scope** | Entire session | Triggered per task |\n| **Location** | `context_modules/personas/` | `skills/*/SKILL.md` |\n| **Activation** | Loaded at session start | Loaded on-demand by trigger phrases |\n| **Registered in `.skills.json`?** | No | Yes |\n| **Contains rules?** | References rules via `@rule:` | May embed rules inline |\n\nA persona provides the stable identity and preferences for a session. Skills provide the domain-specific execution instructions for individual tasks. They complement each other and are both loaded alongside the constitution.\n\n---\n\n## Skills\n\nSkills are self-contained, reusable AI agent capabilities. Each skill packages a domain's instructions, context, and optional automation so an AI agent can reliably apply it on demand without requiring the user to supply background knowledge each time.\n\n### How Skills Work\n\nWhen a user makes a request, an agent:\n\n1. Reads `.skills.json` to discover available skills.\n2. Matches the user's intent against each skill's `description` and trigger phrases.\n3. Loads the relevant `SKILL.md` to obtain domain-specific instructions.\n4. Optionally loads files from `references/` for deeper context.\n5. Applies the skill while completing the task.\n\nSkills are loaded **on-demand** — only the skills relevant to the current request are activated. Default skills listed in `.skills.json` are auto-installed into the agent's skills directory during project init.\n\n### Skill Folder Structure\n\nAll internal skills live under `skills/{skill-name}/`:\n\n```\nskills/\n└── my-skill/\n    ├── SKILL.md          # Required — primary instructions for the agent\n    ├── references/       # Optional — supplementary reference documents\n    │   ├── guide.md\n    │   └── patterns.md\n    └── scripts/          # Optional — automation scripts for the skill\n        └── setup.sh\n```\n\n#### `SKILL.md`\n\nThe entry point for every skill. It uses a YAML front matter block to declare metadata, followed by Markdown instructions:\n\n```yaml\n---\nname: my-skill\ndescription: >\n  Short description of the skill's purpose.\n  Use when [trigger phrases that describe when to activate this skill].\n---\n\n# My Skill\n\n## What This Skill Provides\n...\n\n## When to Use This Skill\n...\n\n## Core Patterns\n...\n```\n\nThe `description` field is critical — it tells the agent **when** to activate the skill. Write it in natural language and include representative phrases a user might say.\n\n### The `.skills.json` Manifest\n\n`.skills.json` is the single source of truth for skill discovery and policy. It defines which local skills are auto-installed during project init, which external skills are available on demand, and which skills are blocked.\n\n```json\n{\n  \"version\": \"2.0.0\",\n  \"source\": \"team-ai-directives\",\n  \"description\": \"Team skills manifest. The `default` list contains skill names that are auto-installed during project init. The `external` map contains on-demand skills fetched by URL. The `blocked` list contains skills that must never be installed.\",\n  \"default\": [ ... ],\n  \"external\": { ... },\n  \"blocked\": [ ... ],\n  \"policy\": { ... }\n}\n```\n\n#### Skill Categories in `.skills.json`\n\n| Category | Meaning |\n|---|---|\n| `default` | Local skills (from this repository's `skills/` directory) auto-installed into the agent's skills directory during project init |\n| `external` | Skills fetched on demand from a URL; not stored locally |\n| `blocked` | Skills explicitly prohibited; the agent must refuse to use them |\n\n#### Skill Entry Format\n\n**Local skills** are listed by folder name in the `default` array:\n\n```json\n\"default\": [\n  \"dbt-template\",\n  \"github-actions\",\n  \"helm-charts\"\n]\n```\n\n**External skills** are keyed by name with metadata pointing to the raw `SKILL.md`:\n\n```json\n\"external\": {\n  \"react-best-practices\": {\n    \"version\": \"^1.0.0\",\n    \"description\": \"...\",\n    \"categories\": [\"frontend\", \"react\"],\n    \"source\": \"https://github.com/org/repo\",\n    \"url\": \"https://raw.githubusercontent.com/org/repo/main/skills/skill-name/SKILL.md\"\n  }\n}\n```\n\n### Configuring Skills\n\n#### Adding a Local Skill\n\n1. Create the skill folder and `SKILL.md` (see Creating a New Skill below).\n2. Add the skill's folder name to the `default` array in `.skills.json`:\n\n```json\n\"default\": [\n  \"dbt-template\",\n  \"my-skill\"\n]\n```\n\n#### Adding an External Skill\n\nExternal skills are fetched from a URL at runtime. Add them to the `external` map:\n\n```json\n\"external\": {\n  \"my-external-skill\": {\n    \"version\": \"^1.0.0\",\n    \"description\": \"Short description with trigger phrases.\",\n    \"categories\": [\"relevant\", \"tags\"],\n    \"source\": \"https://github.com/org/repo\",\n    \"url\": \"https://raw.githubusercontent.com/org/repo/main/skills/skill-name/SKILL.md\"\n  }\n}\n```\n\n#### Blocking a Skill\n\nTo prevent an agent from using a specific skill (e.g., a deprecated or insecure external skill), add it to the `blocked` list:\n\n```json\n\"blocked\": [\n  {\n    \"id\": \"github:unsafe-org/deprecated-skill\",\n    \"reason\": \"Security vulnerability - deprecated by maintainer\"\n  }\n]\n```\n\n### Creating a New Skill\n\n```bash\nmkdir -p skills/my-skill/references\n```\n\nCreate `skills/my-skill/SKILL.md`:\n\n```yaml\n---\nname: my-skill\ndescription: >\n  Describe what the skill does. Use when the user asks to [action],\n  [another action], or [trigger phrase].\n---\n\n# My Skill\n\n## What This Skill Provides\n\nBrief overview of the domain knowledge and capabilities this skill covers.\n\n## When to Use This Skill\n\n- Scenario 1\n- Scenario 2\n\n## Core Patterns\n\n### Pattern Name\n\n**Rule**: State the rule clearly.\n\n**Implementation**:\n- Step or detail\n- Step or detail\n\n**References**: See references/guide.md\n```\n\nThen register it in `.skills.json`.\n\n### Using Skills as an Agent\n\nWhen processing a request, an agent resolves skills in this order:\n\n1. **Constitution** — `context_modules/constitution.md` (always loaded).\n2. **Persona** — relevant file from `context_modules/personas/` based on task context.\n3. **Skill** — triggered by matching the user's intent to a skill description.\n\nTo activate a skill manually, tell the agent which skill to use:\n\n> \"Using the `github-actions` skill, create a reusable workflow for deploying to Kubernetes.\"\n\nThe agent will read `skills/github-actions/SKILL.md` and any referenced rule files before responding.\n\n### How Personas and Skills Work Together\n\nA persona and one or more skills are loaded at the same time. They complement rather than duplicate each other.\n\nFor example, a DevOps Engineer session might look like:\n\n1. **Constitution** — foundational team principles always apply\n2. **Persona**: `devops_engineer.md` — sets the role identity, collaboration preferences, and tool context\n3. **Skill**: `github-actions` — activated when the user asks about CI/CD pipelines\n4. **Skill**: `helm-charts` — activated when the user asks about packaging for Kubernetes\n\nThe persona tells the agent _who it is_. The skills tell it _how to execute_ specific tasks.\n\n### External Skills\n\nExternal skills are fetched at runtime from their `url` field in `.skills.json`. They are not stored locally in this repository.\n\nTo discover external skills, ask your AI agent:\n\n> \"What skills are available in the external registry?\"\n\nThe agent will read `.skills.json`, list the `external` entries, and describe when each is useful. To use one, the agent fetches the `SKILL.md` from the provided `url`.\n\n### Policy Settings\n\nThe `policy` section of `.skills.json` controls agent behavior:\n\n```json\n\"policy\": {\n  \"auto_install_default\": true,\n  \"enforce_blocked\": true,\n  \"allow_project_override\": true\n}\n```\n\n| Setting | Default | Description |\n| --- | --- | --- |\n| `auto_install_default` | `true` | Skills in the `default` list are automatically installed during project init |\n| `enforce_blocked` | `true` | The agent refuses to use any skill in the `blocked` list |\n| `allow_project_override` | `true` | Individual projects can override manifest settings locally |\n\n---\n\n## Governance Commands and Skills\n\nGovernance capabilities are available through both delivery mechanisms — as **spec-kit commands** from the bundled `team-ai-directives` extension (installed by `specify init --team-ai-directives <this-repo>`) and as **agent skills** from [adlc-team-skills](https://github.com/tikalk/adlc-team-skills):\n\n| Capability | Spec Kit command | Agent Skill | Purpose |\n|---|---|---|---|\n| Bootstrap session | `team.boot` | `team-boot` (auto) | Load the constitution and orient the agent before any task |\n| Discover context | `team.discover` | `team-discover` (auto) | Find relevant personas, rules, examples, and skills for the current task |\n| Set up knowledge base | `specify init --team-ai-directives` | `team-setup` | Clone, point at, or scaffold the knowledge base |\n| Repair | `team.repair` | `team-repair` | Re-index CDR.md, .skills.json, and AGENTS.md; health check; conflict scan; freshness verification |\n| Manage skills | `team.skills` | `team-skills` | Browse and install team skills from the knowledge base |\n| Verify health | `team.verify` | `team-repair --health-only` | Verify knowledge base config, skills registry, CDR tracking, and constitution alignment |\n| Curate CDRs | `team.curate`, `levelup.init` / `levelup.specify` | `levelup-init` / `levelup-specify` | Propose Context Directive Records from a codebase or completed feature |\n| Review CDRs | `levelup.clarify` | `levelup-clarify` | Accept, reject, or defer proposed CDRs |\n| Publish CDRs | `team.evolve`, `levelup.implement` | `levelup-implement` | Compile accepted CDRs into knowledge base artifacts and a draft PR |\n| Validate | `levelup.validate` | `team-repair --conflicts` / `--freshness` | Scan for rule conflicts and update verification timestamps |\n\n**Naming conventions:** spec-kit commands are canonically named `adlc.team-ai-directives.*` / `adlc.levelup.*` and are invoked via the short aliases shown above (`team.discover`, `levelup.init`). Agent skills use dash-names (`team-discover`, `levelup-init`) and are model-invoked through the [Agent Skills standard](https://agentskills.io).\n\n### Integration\n\nIn **spec-kit projects**, the `agent-context` extension injects team-directives awareness into the project's context file during `specify init`. It prompts the agent to invoke `team.discover` before feature work and to inherit the team constitution when updating project principles. Run verification anytime by invoking the `team.verify` command.\n\nIn **skills-based projects**, `team-boot` performs the equivalent role — loading the constitution at session start and chaining into `team-discover`. Run health checks anytime with the `team-repair` skill.\n\n---\n\n## Versioning\n\nUse git tags (v1.0.0, v2.0.0, etc.) to manage breaking changes. Downstream consumers can pin to specific versions.\n\n## Customization\n\nFork this repository and customize:\n\n1. **Constitution** - Add your team's principles\n2. **Personas** - Define roles matching your team\n3. **Rules** - Add domain-specific patterns\n4. **Skills** - Create capabilities for your workflows\n5. **`.mcp.json`** - Configure your MCP servers\n6. **`.skills.json`** - Register your skills\n\n## Contributing\n\nSee [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines. All changes require pull request review.\n",
  "bytes": 25611,
  "sha": "9892e6a0a46bd9d65f04dc9e4ef5c0845d3d94c989dd096f04a1e68587db4b9a",
  "repo_slug": "tikalk/agentic-sdlc-team-ai-directives",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/okf_tikalk_agentic_sdlc_team_ai_directives_c_7b2ed4db/readme"
}