### Install and Initialize APM CLI Source: https://github.com/danielmeppiel/agentic-sdlc-handbook/blob/main/past_reviews/PANEL-SYNTHESIS-FINAL.md This snippet shows the command-line interface commands for installing and initializing the APM tool. It is presented as a call-to-action in the documentation. ```bash pip install apm-cli && apm init ``` -------------------------------- ### APM Configuration File Source: https://github.com/danielmeppiel/agentic-sdlc-handbook/blob/main/handbook/reviews/apm-final-report.md Example of an APM configuration file (`apm.yml`) used for project setup and installation. ```yaml apm.yml ``` -------------------------------- ### Multi-Tool Example - Copilot Output Source: https://github.com/danielmeppiel/agentic-sdlc-handbook/blob/main/handbook/reviews/apm-final-report.md Compiled output for GitHub Copilot from source primitives. ```unknown compiled-copilot/ ``` -------------------------------- ### Multi-Tool Example - Source Primitives Source: https://github.com/danielmeppiel/agentic-sdlc-handbook/blob/main/handbook/reviews/apm-final-report.md Canonical source primitives used for multi-tool compilation examples. ```unknown source/ ``` -------------------------------- ### Agent Configuration YAML Example Source: https://github.com/danielmeppiel/agentic-sdlc-handbook/blob/main/handbook/reviews/ch11-dev-lead-proxy.md A sample YAML configuration for an agent, demonstrating settings for tools, model, and other parameters. ```yaml tools: - name: "code_interpreter" description: "Execute Python code." module: "agents.tools.code_interpreter" - name: "file_finder" description: "Find files in the workspace." module: "agents.tools.file_finder" model: "claude-3-sonnet-20240229" # ... other agent configuration ... ``` -------------------------------- ### Copilot Instructions for APM Project Source: https://github.com/danielmeppiel/agentic-sdlc-handbook/blob/main/handbook/reviews/apm-final-report.md Example of Copilot instructions within a starter project configured with APM. ```markdown .github/copilot-instructions.md ``` -------------------------------- ### Multi-Tool Example - Cursor Output Source: https://github.com/danielmeppiel/agentic-sdlc-handbook/blob/main/handbook/reviews/apm-final-report.md Compiled output for Cursor, including .mdc files, from source primitives. ```unknown compiled-cursor/ ``` -------------------------------- ### Instructions (.instructions.md) - Eager Preload Source: https://context7.com/danielmeppiel/agentic-sdlc-handbook/llms.txt Scoped conventions loaded automatically when an agent touches paths matching the `applyTo` glob. Keep files under 40-50 lines; longer files dilute attention. ```markdown --- applyTo: "src/payments/**" description: "Payment service conventions — currency handling and idempotency" --- # Payments Module Rules ## Currency - Always use `Decimal`, never `float`, for monetary amounts. - Store amounts in the smallest currency unit (cents). Convert at display layer only. ## Idempotency - Every POST endpoint requires an `Idempotency-Key` header. - Duplicate requests return the original response; never re-process. ## Error Codes - Use payment-specific error codes from `payments/errors.py`. See [payment error taxonomy](../../docs/payments/errors.md) for the full list. ``` -------------------------------- ### Companion Package Structure Source: https://github.com/danielmeppiel/agentic-sdlc-handbook/blob/main/handbook/reviews/apm-final-report.md Defines the directory structure for the companion package, including APM-specific and manual project setups, as well as multi-tool examples. ```tree prose-handbook-companion/ ├── apm.yml # Makes it installable via apm install ├── README.md # Explains both usage paths (clone vs install) ├── instructions/ │ └── prose-conventions.md # PROSE methodology as agent instructions ├── agents/ │ └── prose-reviewer.agent.md # PROSE compliance reviewer ├── prompts/ │ └── architecture-review.md # Reusable prompt from Chapter 8 ├── templates/ │ └── primitive-inventory.md # Lightweight inventory template ├── examples/ │ ├── starter-project-with-apm/ # For APM users │ │ ├── apm.yml │ │ └── .github/ │ │ └── copilot-instructions.md │ ├── starter-project-manual/ # For non-APM users (CTO request) │ │ ├── README.md # Explains manual workflow step by step │ │ └── .github/ │ │ ├── copilot-instructions.md │ │ └── instructions/ │ │ └── example-convention.instructions.md │ └── multi-tool/ # Same primitives → 3 tool formats (Dev Lead request) │ ├── source/ # Canonical primitives │ ├── compiled-copilot/ # Output for GitHub Copilot │ ├── compiled-cursor/ # Output for Cursor (.mdc files) │ └── compiled-claude-code/ # Output for Claude Code (CLAUDE.md) ``` -------------------------------- ### Manual Project Instructions Source: https://github.com/danielmeppiel/agentic-sdlc-handbook/blob/main/handbook/reviews/apm-final-report.md Instructions for setting up a project manually, without APM, detailing each step. ```markdown .github/instructions/example-convention.instructions.md ``` -------------------------------- ### Mermaid: Governance Readiness Decision Aid Source: https://github.com/danielmeppiel/agentic-sdlc-handbook/blob/main/handbook/reviews/visual-audit-block1.md A decision-making flowchart for governance readiness, guiding users on where to start investments based on their current maturity level. Use this to determine initial governance focus areas. ```mermaid flowchart TD START["Assess 6
governance capabilities"] --> Q1{"Any capability
at 'None'?"} Q1 -->|Yes| FIX["Start here:
Audit Trails +
Access Controls"] Q1 -->|No| Q2{"All at
'Basic' or above?"} Q2 -->|Yes| EXPAND["Safe to expand
agent adoption"] Q2 -->|No| FIX FIX --> REASSESS["Reassess
quarterly"] EXPAND --> MATURE["Invest toward
'Enterprise' per
regulatory scope"] REASSESS --> Q1 style FIX fill:#e74c3c,color:#fff style EXPAND fill:#2ecc71,color:#fff style MATURE fill:#3498db,color:#fff ``` -------------------------------- ### Install APM Dependencies Source: https://github.com/danielmeppiel/agentic-sdlc-handbook/blob/main/packages/editorial-infographic/README.md Run this command after updating your `apm.yml` to install the necessary agents and skills. ```bash apm install ``` -------------------------------- ### CLI Logging UX Example Source: https://github.com/danielmeppiel/agentic-sdlc-handbook/blob/main/handbook/reviews/ch11-dev-lead-proxy.md Illustrates a command-line interface logging user experience, showcasing a decision framework versus a simple rule list. ```bash # Example CLI interaction $ my-agent --log-level DEBUG --output-format json # Agent starts, logs detailed debug information in JSON format. ``` -------------------------------- ### Task Decomposition Example Source: https://context7.com/danielmeppiel/agentic-sdlc-handbook/llms.txt Illustrates how to break down large tasks into smaller, manageable sessions for agents. Each session should have a clear deliverable and fresh context. ```text # Examples of tasks that were too big — and how they got split: TOO BIG: "Add JWT authentication." SPLIT: Five sessions — token schema, middleware, refresh endpoint, tests, frontend. Each with fresh context and a single deliverable. TOO BIG: "Fix the login bug and update the session tests." SPLIT: Bug fix in one session; test updates in follow-up with fix already committed. TOO BIG: "Refactor the payment service to use the new API client." SPLIT: By call site — each session migrates one module, with only the old→new API mapping and the target module in context. ``` -------------------------------- ### Agent Package Manager Install Command Source: https://context7.com/danielmeppiel/agentic-sdlc-handbook/llms.txt Demonstrates the 'apm install' command for resolving and pinning dependencies, generating an 'apm.lock.yaml' file. ```bash # Install resolves the dependency graph and pins every node to a content hash apm install # → resolves code-review-rubric@v1.2 (sha256: 9d10...) # → resolves style-guide-python@v2.0 (sha256: 4b87...) # → writes apm.lock.yaml # Now: one postmortem sharpens the rubric # edit code-review-rubric/SKILL.md, bump to v1.3 # apm install in both consumers picks up v1.3 # both skills produce the same verdict on Monday morning ``` -------------------------------- ### APM - Install and Compile Agents Source: https://context7.com/danielmeppiel/agentic-sdlc-handbook/llms.txt Commands to manage agent dependencies using APM. \'apm install\' resolves and pins dependencies to a lockfile, ensuring reproducibility. \'apm compile\' processes agent instructions into a usable format. \'apm install [package]\' installs a specific skill package. ```bash # Install all declared agent dependencies (resolves and pins to lockfile) apm install # Compile agent instructions (resolves PROSE specs into harness-ready form) apm compile # Install a specific skill package apm install danielmeppiel/genesis ``` -------------------------------- ### Install and Invoke Genesis Skill Source: https://context7.com/danielmeppiel/agentic-sdlc-handbook/llms.txt Commands to install the Genesis skill using 'apm' and invoke it within an agent session for task architecture design. ```bash # Install the Genesis skill apm install danielmeppiel/genesis # Invoke in your agent session (works across all major harnesses) /genesis # Example invocations: /genesis add authentication to this FastAPI service /genesis refactor the payment module to use the Strategy pattern /genesis set up an agent review panel for this repository # Works with: GitHub Copilot, Claude Code, Cursor, Codex, OpenCode # Inspectable at: https://github.com/danielmeppiel/genesis ```