### Get Self-Documenting Explanation

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/SKILL.md

Use `quickstart` with `--json` to get a self-documenting explanation of the system.

```bash
# Self-documenting explanation
cm quickstart --json
```

--------------------------------

### Quick Capability Check and Self-Explanation

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Use the `quickstart` command to get a complete explanation of the system and how to use it. The output is in JSON format.

```bash
# Quick capability check and self-explanation
cm quickstart --json
# → Returns complete explanation of the system and how to use it
```

--------------------------------

### Install Script Options

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Examples of using the install script with specific options like version, destination, system-wide install, easy mode, and verification.

```bash
# Full options
install.sh [--version vX.Y.Z] [--dest DIR] [--system] [--easy-mode] [--verify]
           [--from-source] [--quiet]

# Examples
install.sh --version v0.2.2 --verify    # Specific version
install.sh --system --verify             # Install to /usr/local/bin (requires sudo)
install.sh --from-source                 # Build from source (requires bun)
```

--------------------------------

### Install from Source

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/SKILL.md

Install the Claude Memory System from source by cloning the repository, installing dependencies, building, and moving the executable.

```bash
git clone https://github.com/Dicklesworthstone/cass_memory_system.git
cd cass_memory_system
bun install && bun run build
sudo mv ./dist/cass-memory /usr/local/bin/cm
```

--------------------------------

### Install cass-reflect CLI

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/claude_version/README.md

Install the cass-reflect tool using Bun. This includes installing dependencies and running the CLI.

```bash
bun install

bun run cass-reflect.ts --help

bun run build
./dist/cass-reflect --help
```

--------------------------------

### Install System-Wide

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/SKILL.md

Install the Claude Memory System system-wide using the install script with the '--system' flag.

```bash
install.sh --system --verify
```

--------------------------------

### Install Dependencies

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/CONTRIBUTING.md

Install project dependencies using Bun. Ensure Bun is installed and updated.

```bash
bun install
```

--------------------------------

### Installation - One-Liner

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

The recommended one-liner command for installing the Cass Memory System CLI.

```bash
# Recommended One-Liner (Installation)
```

--------------------------------

### Get Context for Task CLI Example

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/docs/planning/AGENT_FRIENDLINESS_REPORT.md

Example of how to use the 'context' command with the '--json' flag to retrieve task-specific information.

```bash
cm context "implement feature X" --json
```

--------------------------------

### Clone Repository and Install Dependencies

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Clone the cass-memory system repository and install its dependencies using bun.

```bash
git clone https://github.com/Dicklesworthstone/cass_memory_system.git
cd cass_memory_system
bun install
```

--------------------------------

### Get context before starting a task

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Generates context for a given task description. Use the --json flag for machine-readable output.

```bash
cm context "implement user authentication" --json
```

--------------------------------

### Build from Source

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Clone the repository, install dependencies with bun, build the project, and move the binary to the system path.

```bash
git clone https://github.com/Dicklesworthstone/cass_memory_system.git
cd cass_memory_system
bun install
bun run build
sudo mv ./dist/cass-memory /usr/local/bin/cm
```

--------------------------------

### One-liner Install (Linux/macOS)

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Installs cass-memory using curl and bash. The --verify flag ensures the integrity of the installation.

```bash
curl -fsSL "https://raw.githubusercontent.com/Dicklesworthstone/cass_memory_system/main/install.sh?$(date +%s)" \
  | bash -s -- --easy-mode --verify
```

--------------------------------

### Project Command Examples

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/docs/planning/PLAN_FOR_CASS_MEMORY_SYSTEM.md

Export playbook content into agent-specific formats for documentation. Supports various formats and options for customization.

```bash
cm project --format agents.md --output ./AGENTS.md
```

```bash
cm project --format claude.md --per-category 10
```

```bash
cm project --show-counts
```

--------------------------------

### Install cass CLI

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/claude_version/README.md

Install the cass CLI tool using a curl script. This is a prerequisite for using cass-reflect.

```bash
curl -fsSL https://raw.githubusercontent.com/Dicklesworthstone/coding_agent_session_search/main/install.sh | bash
```

--------------------------------

### Show Playbook Stats CLI Example

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/docs/planning/AGENT_FRIENDLINESS_REPORT.md

Example of the 'stats' command with '--json' to display playbook health metrics.

```bash
cm stats --json
```

--------------------------------

### Install via Scoop (Windows)

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Installs cass-memory using the Scoop package manager. Requires adding the custom bucket first.

```bash
scoop bucket add dicklesworthstone https://github.com/Dicklesworthstone/scoop-bucket
scoop install dicklesworthstone/cm
```

--------------------------------

### Reset Onboarding Process

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/SKILL.md

Use `cm onboard reset` to start the onboarding process fresh.

```bash
cm onboard reset                              # Start fresh
```

--------------------------------

### Record Feedback CLI Example

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/docs/planning/AGENT_FRIENDLINESS_REPORT.md

Example of the 'mark' command used with '--json' to record feedback on a specific bullet.

```bash
cm mark b-xxx helpful --json
```

--------------------------------

### Find Similar Rules CLI Example

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/docs/planning/AGENT_FRIENDLINESS_REPORT.md

Example of using the 'similar' command with '--json' to find rules related to a specific query.

```bash
cm similar "error handling" --json
```

--------------------------------

### Initialize Playbook with Starter

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Command to initialize a new playbook using a specified starter. For example, initializing with the 'typescript' starter.

```bash
# Initialize with a starter
cm init --starter typescript
```

--------------------------------

### ACE Protocol: Learning Command Example

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/2025-12-07-gemini-gemini_-_direct_access_to_google_ai.md

An example of the `ace learn` command used in the 'POST-FLIGHT' phase of the ACE Protocol. This command takes the path to a session log file for validation.

```bash
ace learn <path_to_log>
```

--------------------------------

### Install via Homebrew (macOS/Linux)

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Installs cass-memory using the Homebrew package manager. Ensure you have added the custom tap first.

```bash
brew install dicklesworthstone/tap/cm
```

--------------------------------

### Install Claude Memory System (One-liner)

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/SKILL.md

Install the Claude Memory System using a recommended one-liner curl command. Includes options for easy mode and verification.

```bash
curl -fsSL https://raw.githubusercontent.com/Dicklesworthstone/cass_memory_system/main/install.sh \
  | bash -s -- --easy-mode --verify
```

--------------------------------

### Query cass for diverse examples of successes/failures

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/2025-12-07-grok-integrating_cass_into_reflective_workflows_shared_grok_conversation.md

Use this command to fetch diverse examples of successes or failures from cass for query analysis. Supports wildcard fallback and explanation.

```bash
cass search "*failure* parallelization" --robot --wildcard-fallback --explain for query analysis
```

--------------------------------

### Update CLAUDE.md with Networking Best Practices

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/2025-12-07-gemini-gemini_-_direct_access_to_google_ai.md

Update the CLAUDE.md file with empirical findings. This example suggests using 'httpx' over 'requests' for networking and provides a search query for related examples.

```bash
CLAUDE.md with: **"NETWORKING: Always use `httpx` instead of `requests`. See `cass search 'httpx migration'` for examples."**
```

--------------------------------

### Retrieve Examples with Cass Expand in ACE Reflector

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/2025-12-07-grok-integrating_cass_into_reflective_workflows_shared_grok_conversation.md

After trajectory, retrieve examples for context or analysis. Use cass's fuzzy fallback for broader matches. Reflect on combined input to extract lessons.

```bash
cass expand /path/to/session.jsonl -n 42 -C 3 --json
```

```bash
cass search "*failure* code quality" --robot --explain
```

--------------------------------

### Install Dependencies with Bun

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/2025-12-07-gemini-gemini_-_direct_access_to_google_ai.md

Install necessary dependencies for the ace.ts CLI using Bun. This includes Commander for CLI argument parsing, Zod for schema validation, and the Vercel AI SDK with OpenAI support.

```bash
bun add commander zode ai @ai-sdk/openai zod
```

--------------------------------

### List available starter templates

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Lists the available starter templates for initialization.

```bash
cm starters
```

--------------------------------

### Build Playbook (Onboarding Sample with Fill Gaps)

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Generates a sample playbook onboarding entry, filling in any gaps. Use --json for structured output.

```bash
cm onboard sample --fill-gaps --json
```

--------------------------------

### Get Task-Specific Memory

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Retrieves memory context relevant to a specific task before starting. Always use the --json flag in agent contexts.

```bash
cm context "implement auth rate limiting" --json
```

--------------------------------

### Retrieve Context for Tasks with cm

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/AGENTS.md

Before starting complex tasks, retrieve relevant context using the `cm context` command. This command helps gather potential rules, anti-patterns, past session examples, and suggested search queries for deeper investigation.

```bash
cm context "<task description>" --json
```

--------------------------------

### Apply Starter Playbook to Existing Project

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/SKILL.md

Use 'cm playbook bootstrap <starter_name>' to apply a starter playbook to an existing project.

```bash
cm playbook bootstrap react
```

--------------------------------

### Initialize configuration and playbook

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Initializes the system configuration and playbook. Can use starter templates.

```bash
cm init
```

```bash
cm init --starter typescript
```

--------------------------------

### List Available Starters

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Command to list all available starter playbooks. This is the first step in initializing or applying a starter playbook.

```bash
# List all available starters
cm starters
```

--------------------------------

### Initialize Cass Memory Project

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/SKILL.md

Use 'cm init' to initialize a new project. Specify a starter template with '--starter' or reinitialize with '--force' to create a backup.

```bash
cm init
```

```bash
cm init --starter typescript
```

```bash
cm init --force
```

--------------------------------

### Apply Starter to Existing Playbook

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Command to apply a starter playbook to an existing playbook. This is useful for bootstrapping a project with a specific tech stack's best practices.

```bash
# Apply a starter to an existing playbook
cm playbook bootstrap react
```

--------------------------------

### Bootstrap Playbook with Starter Rules

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/SKILL.md

Apply starter rules for a specific technology using `cm playbook bootstrap`.

```bash
cm playbook bootstrap react                   # Apply starter to existing
```

--------------------------------

### Install Specific Version

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/SKILL.md

Install a specific version of the Claude Memory System using the install script with the '--version' flag.

```bash
install.sh --version v0.2.2 --verify
```

--------------------------------

### Install Trauma Guard Hooks

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/SKILL.md

Install Trauma Guard hooks using `cm guard`. Use `--install` for the Claude Code hook, `--git` for a Git pre-commit hook, or both.

```bash
cm guard --install                            # Claude Code hook
```

```bash
cm guard --git                                # Git pre-commit hook
```

```bash
cm guard --install --git                      # Both
```

--------------------------------

### Check installation status of safety hooks

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Checks the installation status of the safety hooks.

```bash
cm guard --status
```

--------------------------------

### List Cass Memory Starters

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/SKILL.md

Use 'cm starters' to list available starter templates for project initialization or playbook bootstrapping.

```bash
cm starters
```

--------------------------------

### Markdown Integration Example

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/claude_version/README.md

Example of how to integrate Claude Code or Codex into a CLAUDE.md file.

```markdown
Add to your `CLAUDE.md`:

```markdown

```

--------------------------------

### Diary Output Example

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/claude_version/README.md

Example JSON output structure for a generated diary entry.

```json
{
  "sessionPath": "/home/user/.claude/projects/myapp/session-123.jsonl",
  "timestamp": "2025-12-07T10:30:00Z",
  "agent": "claude_code",
  "workspace": "/projects/myapp",
  "accomplishments": ["Implemented JWT refresh endpoint", "Added rate limiting"],
  "decisions": ["Chose sliding window for rate limiting over fixed window"],
  "challenges": ["Redis connection pooling caused timeout issues"],
  "preferences": ["User prefers explicit error messages over codes"],
  "keyLearnings": ["Always set maxRetriesPerRequest on ioredis client"]
}
```

--------------------------------

### Manage Configuration

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/claude_version/README.md

Commands for showing, setting, and locating configuration settings.

```bash
# Show config
cass-reflect config show
```

```bash
# Update settings
cass-reflect config set provider openai
```

```bash
cass-reflect config set model gpt-4o
```

```bash
cass-reflect config set maxReflectorIterations 5
```

```bash
# Show config path
cass-reflect config path
```

--------------------------------

### Check Onboarding Status and Recommendations

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Checks the current onboarding status and provides recommendations for next steps. This is a good starting point for the onboarding process.

```bash
cm onboard status
```

--------------------------------

### Playbook Schema Example

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/claude_version/README.md

Example YAML structure for a playbook entry, including metadata and bullet details.

```yaml
schema_version: 1
name: "my-project"
description: "Accumulated coding wisdom"

metadata:
  last_reflection: "2025-12-07T10:00:00Z"
  total_reflections: 42
  created_at: "2025-11-01T00:00:00Z"

bullets:
  - id: "b-1a2b3c4d-x7y8"
    category: "debugging"
    content: |
      When debugging TypeScript type errors with generics, 
      use `tsc --noEmit --pretty` and check the instantiated 
      type at the call site, not just the generic definition.
    created_at: "2025-11-15T14:30:00Z"
    updated_at: "2025-12-05T09:00:00Z"
    helpful_count: 8
    harmful_count: 1
    tags: ["typescript", "debugging", "generics"]
    source_sessions:
      - "/home/user/.claude/projects/app/session-001.jsonl:142"
      - "/home/user/.cursor/state.vscdb:session-xyz"
```

--------------------------------

### Install Trauma Guard Hooks via CLI

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

These bash commands install the necessary hooks for the Trauma Guard system to intercept and prevent dangerous commands. Options include installing the Claude Code hook, the Git pre-commit hook, or both.

```bash
# Install Claude Code hook (intercepts Bash commands)
cm guard --install
```

```bash
# Install Git pre-commit hook (scans commits)
cm guard --git
```

```bash
# Install both
cm guard --install --git
```

```bash
# Check installation status
cm guard --status
```

--------------------------------

### Install safety hooks

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Installs safety hooks for various environments like Claude Code or Git pre-commit.

```bash
cm guard --install
```

```bash
cm guard --git
```

```bash
cm guard --install --git
```

--------------------------------

### Deprecate Bullet Delta Example

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/gpt_pro_version.md

An example of a 'deprecate_bullet' delta, used to mark a playbook entry as outdated or harmful.

```json
{
  "type": "deprecate_bullet",
  "id": "cass-global-0012"
}
```

--------------------------------

### Context Command Example with JSON Output

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Demonstrates the `context` command with a specific task and the expected structured JSON output, including relevant bullets, anti-patterns, and history snippets.

```bash
cm context "fix the auth timeout bug" --json
```

--------------------------------

### Cass CLI Commands for Presets

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/docs/planning/SUGGESTED_IMPROVEMENTS_TO_CASS_BASED_ON_CMS.md

Demonstrates how to add SSH sources with predefined platform-specific default paths using presets.

```bash
cass sources add ssh work-laptop --host work-laptop --preset macos-defaults
```

```bash
cass sources add ssh home-server --host home-server --preset linux-defaults
```

--------------------------------

### Get Specific Rule Details

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/SKILL.md

Use `cm playbook get` followed by a rule ID to retrieve its details.

```bash
cm playbook get b-8f3a2c                      # Rule details
```

--------------------------------

### Initialize CLI and Imports

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/2025-12-07-grok-integrating_cass_into_reflective_workflows_shared_grok_conversation.md

Sets up the command-line interface using commander and imports necessary modules for AI interaction, file system operations, and process spawning. Ensure all dependencies are installed.

```typescript
#!/usr/bin/env bun

import { Command } from 'commander';
import { openai } from '@ai-sdk/openai'; // Or import { anthropic } from '@ai-sdk/anthropic';
import { generateText } from 'ai'; // Vercel AI SDK core
import { spawn } from 'bun';
import * as fs from 'fs/promises';
import * as path from 'path';
import { fileURLToPath } from 'url';
```

--------------------------------

### Verify Installation

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Check the installed version of the cm CLI and run a system health check using the doctor command.

```bash
cm --version
cm doctor --json
```

--------------------------------

### Run Development Server (One-off)

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/CONTRIBUTING.md

Execute a command once using the development server. Replace `<command>` and `[args]` with your specific command and arguments.

```bash
bun run dev -- <command> [args]
```

--------------------------------

### Build Playbook (Onboarding Status)

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Checks the status of the playbook onboarding process. Use --json for structured data output.

```bash
cm onboard status --json
```

--------------------------------

### Generate Prioritized Onboarding Samples

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/SKILL.md

Use `cm onboard sample` to generate prioritized sessions for onboarding. Use `--fill-gaps` to prioritize filling gaps. Filter by agent (`--agent`) or time (`--days`).

```bash
cm onboard sample --fill-gaps                 # Prioritized sessions
```

```bash
cm onboard sample --agent claude --days 14    # Filter by agent/time
```

```bash
cm onboard sample --workspace /path/project   # Filter by workspace
```

```bash
cm onboard sample --include-processed         # Re-analyze sessions
```

--------------------------------

### Check Onboarding Progress

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/SKILL.md

Use `cm onboard status` to check the progress of the agent-native playbook building.

```bash
cm onboard status                             # Check progress
```

--------------------------------

### Category Keyword Detection Example

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

A TypeScript example demonstrating how category keywords can be used to detect relevant categories from text content.

```typescript
// Example: detecting categories from text
const CATEGORY_KEYWORDS = {
  debugging: ["debug", "error", "fix", "bug", "trace", "stack", ...],
  testing: ["test", "mock", "assert", "expect", "jest", "vitest", ...],
  security: ["security", "auth", "token", "encrypt", "permission", ...],
  // ... more categories
};

// Count keyword matches → category with most matches wins
```

--------------------------------

### Agent Playbook Section Example

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/gpt_pro_version.md

An example of how playbook rules can be presented in a markdown file for agent documentation, including helpfulness counts.

```markdown
## Agent Playbook (auto-generated from Cass)

These rules are synthesized from past sessions across Claude, Cursor, Codex, and other agents, using Cass + reflection.

### Testing (top 5 rules)

1. [3× helpful] Run `pnpm test` for the current package before committing, and rerun after updating dependencies.
2. [2× helpful] For flaky tests, reproduce in isolation with `pnpm vitest run -t "<test name>"` and document any nondeterminism.
3. [1× helpful] After moving files, always run `pnpm lint --filter <project_name>` to validate TypeScript and ESLint rules.
```

--------------------------------

### Define a Custom Django Starter Playbook (src/)

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Create a YAML file in the src/starters/ directory to define a custom starter playbook for Django. This version includes metadata like version and tags.

```yaml
# src/starters/django.yaml
metadata:
  name: django
  description: Django web framework best practices
  version: "1.0"
  tags: [python, django, web]

bullets:
  - content: "Always use Django's ORM for database operations"
    category: database
    scope: framework
    kind: stack_pattern
    maturity: established

  - content: "Run migrations before deploying"
    category: deployment
    scope: framework
    kind: workflow_rule
    maturity: proven
```

--------------------------------

### Register Agent and Ensure Project

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/AGENTS.md

Register an agent within a project and ensure the project is set up. Use `project_key` as an absolute path.

```python
ensure_project(project_key=<abs-path>)
register_agent(project_key, program, model)
```

--------------------------------

### Update Bullet Delta Example

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/gpt_pro_version.md

An example of an 'update_bullet' delta, used to adjust the helpfulness or harmfulness counts of an existing playbook bullet.

```json
{
  "type": "update_bullet",
  "id": "cass-global-0012",
  "delta_helpful": 1,
  "delta_harmful": 0,
  "note": "Rule was followed and produced a clean PR with minimal feedback."
}
```

--------------------------------

### Handle Cass Unavailability

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/docs/planning/PLAN_FOR_CASS_MEMORY_SYSTEM.md

Checks if cass is available and attempts to find it in common installation paths. If not found, it returns an error with installation instructions.

```typescript
async function handleCassUnavailable(): CommandResult {
  // Check if cass exists
  if (!cassAvailable()) {
    // Try common installation paths
    const commonPaths = [
      "/usr/local/bin/cass",
      "~/.cargo/bin/cass",
      "~/.local/bin/cass"
    ];

    for (const p of commonPaths) {
      if (await fileExists(p)) {
        // Update config with found path
        return { success: true, message: `Found cass at ${p}` };
      }
    }

    return {
      success: false,
      error: "cass not found",
      message: `Install cass from https://github.com/Dicklesworthstone/coding_agent_session_search
               or set CASS_PATH environment variable`
    };
  }
}
```

--------------------------------

### Build Playbook (Read Session with Template)

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/README.md

Reads a session file and generates a playbook entry using a template. Use --json for structured output.

```bash
cm onboard read /path/to/session.jsonl --template --json
```

--------------------------------

### Check Trauma Guard Installation Status

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/SKILL.md

Use `cm guard --status` to check the installation status of the Trauma Guard hooks.

```bash
cm guard --status                             # Check installation
```

--------------------------------

### CLI Setup with Commander.js

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/2025-12-07-grok-integrating_cass_into_reflective_workflows_shared_grok_conversation.md

Initializes a Commander.js CLI application with basic information and defines commands for diary, reflect, curate, and full reflection cycles.

```typescript
// CLI Setup
const program = new Command();
program
  .name('agent-reflect')
  .description('General reflection CLI for coding agents')
  .version('1.0.0');

program.command('diary')
  .description('Generate diary entry')
  .argument('<input>', 'Session trace file')
  .option('-o, --output <file>', 'Output diary file', 'diary.md')
  .option('--cass-query <query>', 'Cass search query')
  .action(async (input, opts) => runDiary(input, opts.output, opts.cassQuery));

program.command('reflect')
  .description('Reflect on diaries')
  .argument('<diaries-dir>', 'Directory of diary files')
  .option('--cass-query <query>', 'Cass search query')
  .option('--json', 'JSON output')
  .action(async (dir, opts) => runReflect(dir, opts.cassQuery, opts.json));

program.command('curate')
  .description('Curate playbook updates')
  .argument('<reflections>', 'Reflections input')
  .option('-p, --playbook <file>', 'Playbook file', 'playbook.md')
  .action(async (refl, opts) => runCurate(refl, opts.playbook));

program.command('full')
  .description('Full reflection cycle')
  .option('--since <days>', 'Process since days ago', '7')
  .option('-o, --output <file>', 'Output playbook', 'playbook.md')
  .option('--limit <num>', 'Max sessions', '10')
  .option('--provider <name>', 'AI provider', 'openai')
  .option('--model <name>', 'AI model')
  .action(async (opts) => runFull(opts.since, opts.output, opts));

program.parseAsync();
```

--------------------------------

### Batch Onboarding Script for Cass-Memory

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/docs/AGENT_NATIVE_ONBOARDING.md

A bash script to automate the agent-assisted onboarding process. It lists sessions to process, and includes placeholders for agent interaction and final checks.

```bash
#!/bin/bash
# Agent-assisted onboarding

# 1. List sessions to process
cass search "*" --robot --limit 50 | jq -r '.hits[].source_path' | sort -u > sessions.txt

# 2. For each session, agent reads and extracts rules
# (Agent does this interactively)

# 3. After extraction, check results
cm playbook list
cm stats --json
```

--------------------------------

### Add Bullet Delta Example

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/gpt_pro_version.md

An example of an 'add_bullet' delta, specifying a new lesson learned with its category, scope, content, and initial counts.

```json
{
  "type": "add_bullet",
  "category": "testing",
  "scope": { "type": "workspace", "path": "/Users/me/devshop" },
  "content": "Before running the full test suite, run targeted tests for files you changed to get faster feedback.",
  "initial_helpful_count": 1,
  "initial_harmful_count": 0,
  "sources": [
    { "diary_path": "...", "session_path": "..." }
  ]
}
```

--------------------------------

### Initialize CLI and Imports for Agent Reflection

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/2025-12-07-grok-integrating_cass_into_reflective_workflows_shared_grok_conversation.md

Sets up the command-line interface using commander and imports necessary modules for AI interaction, file system operations, and process management. Includes type definitions for session traces and CASS hits.

```typescript
#!/usr/bin/env bun

import { Command } from 'commander';
import { openai } from '@ai-sdk/openai'; // Or import { anthropic } from '@ai-sdk/anthropic';
import { generateText } from 'ai'; // Vercel AI SDK core
import { spawn } from 'bun';
import * as fs from 'fs/promises';
import * as path from 'path';
import { fileURLToPath } from 'url';

const __dirname = path.dirname(fileURLToPath(import.meta.url));

interface SessionTrace {
  messages: Array<{ role: string; content: string }>;
  metadata?: { agent?: string; timestamp?: string };
}

interface CassHit {
  agent: string;
  snippet: string;
  score: number;
  path: string;
}

```

--------------------------------

### Get Context with Options

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/claude_version/README.md

Get contextual information for a task with specific options, including workspace path, maximum number of bullets, and JSON output format.

```bash
cass-reflect context "Debug memory leak" --workspace /path/to/project --max-bullets 20 --json
```

--------------------------------

### Get Context for a Task

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/claude_version/README.md

Retrieve relevant playbook entries and historical context for a given task. This is the primary command for agents to call to get contextual information.

```bash
cass-reflect context "Fix the authentication timeout bug in auth.ts"
```

--------------------------------

### Initialize ACE CLI with Commander.js

Source: https://github.com/dicklesworthstone/cass_memory_system/blob/main/competing_proposal_plans/2025-12-07-gemini-gemini_-_direct_access_to_google_ai.md

Sets up the main command-line interface for ACE using the Commander.js library. Includes basic name, description, and version.

```typescript
import { Command } from "commander";

const program = new Command();

program
  .name("ace")
  .description("Agentic Context Engine: Middleware for Agent Memory")
  .version("1.0.0");

```