### Install Mods with Go

Source: https://github.com/charmbracelet/mods/blob/main/README.md

Install the latest version of Mods directly using the Go toolchain. This is a convenient method for Go developers or those who prefer building from source.

```bash
go install github.com/charmbracelet/mods@latest
```

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

### Install Mods on Debian/Ubuntu

Source: https://github.com/charmbracelet/mods/blob/main/README.md

Install Mods on Debian or Ubuntu systems using a custom APT repository. This involves adding the Charm repository and its GPG key.

```bash
sudo mkdir -p /etc/apt/keyrings
curl -fsSL https://repo.charm.sh/apt/gpg.key | sudo gpg --dearmor -o /etc/apt/keyrings/charm.gpg
echo "deb [signed-by=/etc/apt/keyrings/charm.gpg] https://repo.charm.sh/apt/ * *" | sudo tee /etc/apt/sources.list.d/charm.list
sudo apt update && sudo apt install mods
```

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

### Install Mods with Homebrew

Source: https://github.com/charmbracelet/mods/blob/main/README.md

Install Mods on macOS or Linux using the Homebrew package manager. Ensure Homebrew is installed before running this command.

```bash
# macOS or Linux
brew install charmbracelet/tap/mods
```

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

### Install Mods with Nix

Source: https://github.com/charmbracelet/mods/blob/main/README.md

Install Mods using the Nix package manager. This command is suitable for NixOS or users managing their environment with Nix.

```bash
# Nix
nix-shell -p mods
```

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

### Install Mods with Winget

Source: https://github.com/charmbracelet/mods/blob/main/README.md

Install Mods on Windows using the Winget package manager. This command is specific to Windows environments.

```bash
# Windows (with Winget)
winget install charmbracelet.mods
```

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

### Install Mods from FreeBSD Ports

Source: https://github.com/charmbracelet/mods/blob/main/README.md

Install Mods on FreeBSD by compiling from the ports tree. This method is suitable for users who prefer managing software through the ports system.

```bash
cd /usr/ports/sysutils/mods
make install
```

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

### Install Mods with AUR (Arch Linux)

Source: https://github.com/charmbracelet/mods/blob/main/README.md

Install Mods on Arch Linux using the AUR helper 'yay'. This command is specific to Arch Linux distributions.

```bash
# Arch Linux (btw)
yay -S mods
```

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

### Recommend Obscure Shows

Source: https://github.com/charmbracelet/mods/blob/main/examples.md

Get recommendations for obscure shows based on a listing of video files by piping 'ls' output to mods with a recommendation prompt, then to glow.

```bash
ls ~/vids | mods -f "recommend me 10 shows based on these, make them obscure" | glow
```

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

### Install Mods on Fedora/RHEL

Source: https://github.com/charmbracelet/mods/blob/main/README.md

Install Mods on Fedora or RHEL-based systems by adding a custom Yum repository. This method configures the package manager to find the Mods package.

```bash
echo '[charm]
name=Charm
baseurl=https://repo.charm.sh/yum/
enabled=1
gpgcheck=1
gpgkey=https://repo.charm.sh/yum/gpg.key' | sudo tee /etc/yum.repos.d/charm.repo
sudo yum install mods
```

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

### Basic Mods Usage

Source: https://github.com/charmbracelet/mods/blob/main/features.md

The most basic way to use the 'mods' command to get a response for a given prompt.

```bash
mods 'first 2 primes'
```

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

### Continue Specific Conversation with New Title

Source: https://github.com/charmbracelet/mods/blob/main/features.md

Starts a new conversation, then continues a specific previous conversation and saves it with a new title.

```bash
mods --title='naturals' 'first 5 natural numbers'
mods --continue='naturals' --title='naturals.json' 'format as json'
```

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

### Write Documentation Draft

Source: https://github.com/charmbracelet/mods/blob/main/examples.md

Create a draft for new documentation by piping a README file to mods with instructions for new content, then to glow.

```bash
mods "write a new section to this readme for a feature that sends you a free rabbit if you hit r" < README.md | glow
```

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

### Select Specific Models and APIs

Source: https://context7.com/charmbracelet/mods/llms.txt

Choose specific models or API providers using command-line flags. Use `-M` for interactive selection. Supports various providers like OpenAI, Anthropic, Google, Ollama, etc.

```bash
# Use a specific model
mods -m gpt-4 "Explain quantum computing"
```

```bash
# Use model alias
mods -m 4o-mini "Quick question about bash"
```

```bash
# Use a different API provider
mods -a anthropic -m claude-sonnet-4 "Write a haiku about coding"
```

```bash
# Use Ollama for local inference
mods -a ollama -m "llama3.2:3b" "Explain recursion"
```

```bash
# Use Google Gemini
mods -a google -m gemini-2.0-flash-001 "What is machine learning?"
```

```bash
# Interactive model selection
mods -M "My prompt here"
```

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

### View Configuration and Cache Paths

Source: https://context7.com/charmbracelet/mods/llms.txt

Display the locations of the configuration and cache directories used by the 'mods' tool.

```bash
# View configuration and cache paths
mods --dirs
# Output:
# Configuration: /home/user/.config/mods
#         Cache: /home/user/.local/share/mods
```

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

### Recommend Albums (Excluding Soundtracks)

Source: https://github.com/charmbracelet/mods/blob/main/examples.md

Receive album recommendations based on a list of shows, excluding any soundtrack music or music from the shows themselves, by piping 'ls' output to mods, then to glow.

```bash
ls ~/vids | mods -f "recommend me 10 albums based on these shows, do not include any soundtrack music or music from the show" | glow
```

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

### Generate Documentation

Source: https://context7.com/charmbracelet/mods/llms.txt

Convert code (e.g., Go functions) into markdown documentation using Mods. The `-f` flag suggests a formatted output.

```bash
cat api.go | mods -f "Generate API documentation in markdown for these functions"
```

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

### Open Settings Editor

Source: https://context7.com/charmbracelet/mods/llms.txt

Quickly open the 'mods' settings file in your default editor for direct modification.

```bash
# Open settings in your editor
mods --settings
```

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

### Organize Videos by Decade

Source: https://github.com/charmbracelet/mods/blob/main/examples.md

Organize video files by decade and summarize them by piping the output of 'ls' to mods with a specific instruction, then to glow.

```bash
ls ~/vids | mods -f "organize these by decade and summarize each" | glow
```

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

### Manage MCP Servers and Tools

Source: https://context7.com/charmbracelet/mods/llms.txt

Interact with configured MCP servers using command-line flags. List servers, available tools, or disable specific servers for a given request.

```bash
# List configured MCP servers
mods --mcp-list

# List available tools from enabled servers
mods --mcp-list-tools

# Disable specific MCP servers for a request
mods --mcp-disable github "My prompt without GitHub tools"
mods --mcp-disable "*" "Disable all MCP servers"
```

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

### Refactor Go Code

Source: https://github.com/charmbracelet/mods/blob/main/examples.md

Pipe Go source code to mods with an instruction to improve it, then pipe the output to glow for display.

```bash
mods -f "what are your thoughts on improving this code?" < main.go | glow
```

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

### Basic Prompt to Default LLM

Source: https://context7.com/charmbracelet/mods/llms.txt

Send a simple prompt to the default LLM model. The response is rendered with Markdown formatting when output is a TTY.

```bash
# Simple prompt
mods "What are the first 5 prime numbers?"

# Expected output:
# The first 5 prime numbers are: 2, 3, 5, 7, and 11.
```

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

### Explain Error Messages

Source: https://context7.com/charmbracelet/mods/llms.txt

Pipe the standard error output of a command (e.g., `npm run build`) to Mods for explanation and suggested fixes.

```bash
npm run build 2>&1 | mods "Explain this error and suggest fixes"
```

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

### Use Custom Roles and Commands

Source: https://context7.com/charmbracelet/mods/llms.txt

Execute commands using custom roles or list available roles. The --role flag specifies the LLM persona, and input can be piped or provided directly.

```bash
# Use a custom role
mods --role shell "find all .go files modified in the last week"
# Output: find . -name "*.go" -mtime -7

mods --role code-reviewer < mycode.py

# List available roles
mods --list-roles
```

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

### Generate Product Features

Source: https://github.com/charmbracelet/mods/blob/main/examples.md

Generate new feature ideas based on existing source code by piping the code to mods with a specific prompt, then to glow.

```bash
mods -f "come up with 10 new features for this tool." < main.go | glow
```

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

### Generate Test Cases

Source: https://context7.com/charmbracelet/mods/llms.txt

Create unit tests for a given function or script (e.g., Python) by piping the code to Mods.

```bash
cat function.py | mods "Write pytest unit tests for this function"
```

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

### Define Custom Roles for System Prompts

Source: https://context7.com/charmbracelet/mods/llms.txt

Create custom roles in your configuration file to define reusable system prompts for specific use cases. This allows for consistent AI behavior across different interactions.

```yaml
roles:
  developer:
    system_prompt: "You are a helpful AI assistant that writes code."
  writer:
    system_prompt: "You are a creative writer specializing in short stories."
```

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

### Reset Settings to Defaults

Source: https://context7.com/charmbracelet/mods/llms.txt

Revert 'mods' settings to their default values. The current configuration will be backed up before resetting.

```bash
# Reset settings to defaults (backs up old config)
mods --reset-settings
```

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

### Configure MCP Servers for Tool Calling

Source: https://context7.com/charmbracelet/mods/llms.txt

Set up MCP (Model Context Protocol) servers in mods.yml to allow LLMs to interact with external tools. Supports stdio and SSE types.

```yaml
# In ~/.config/mods/mods.yml
mcp-servers:
  github:
    type: stdio
    command: docker
    env:
      - GITHUB_PERSONAL_ACCESS_TOKEN=your_token_here
    args:
      - run
      - "-i"
      - "--rm"
      - "-e"
      - GITHUB_PERSONAL_ACCESS_TOKEN
      - "ghcr.io/github/github-mcp-server"

  filesystem:
    type: stdio
    command: npx
    args:
      - "@anthropic/mcp-server-filesystem"
      - "/path/to/allowed/directory"

  # SSE-based server
  remote-tools:
    type: sse
    url: https://my-mcp-server.example.com/sse

mcp-timeout: 30s  # Timeout for MCP calls (default: 15s)
```

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

### Set Google API Key for Gemini

Source: https://github.com/charmbracelet/mods/blob/main/README.md

Set the GOOGLE_API_KEY environment variable to enable the use of Gemini models from Google.

```sh
export GOOGLE_API_KEY=your-google-api-key
```

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

### Interactive Conversation Workflow

Source: https://context7.com/charmbracelet/mods/llms.txt

Engage in multi-turn AI conversations. Use '--title' to name a conversation, '-C' for subsequent turns, and '--show' to review the full history.

```bash
mods --title="project-plan" "I'm building a CLI tool for task management"
```

```bash
mods -C "What database should I use?"
```

```bash
mods -C "How should I structure the commands?"
```

```bash
mods --show="project-plan"  # Review the full conversation
```

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

### Read Fortune from Downloads

Source: https://github.com/charmbracelet/mods/blob/main/examples.md

Generate a personal fortune based on the files in your downloads folder by piping 'ls' output to mods with a fortune-telling prompt, then to glow.

```bash
ls ~/Downloads | mods -f "tell my fortune based on these files" | glow
```

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

### Set Groq API Key

Source: https://github.com/charmbracelet/mods/blob/main/README.md

Set the GROQ_API_KEY environment variable to utilize models powered by Groq's LPU inference engine.

```sh
export GROQ_API_KEY=your-groq-api-key
```

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

### Process stdin with Mods

Source: https://context7.com/charmbracelet/mods/llms.txt

Pipe content from other commands or files to Mods and ask questions about it. The piped content becomes part of the prompt context. Use this for analyzing code, summarizing API responses, or organizing files.

```bash
# Analyze code and get improvement suggestions
mods -f "What are your thoughts on improving this code?" < main.go
```

```bash
# Summarize API response
curl -s "https://api.open-meteo.com/v1/forecast?latitude=29&longitude=-90&current_weather=true" | \
  mods -f "Summarize this weather data for a human"
```

```bash
# Get recommendations based on file listing
ls ~/Documents | mods -f "Organize these files by category"
```

```bash
# Pipe to other commands for further processing
echo "as JSON" | mods "first 5 prime numbers" | jq .
```

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

### Summarize Weather API Data

Source: https://github.com/charmbracelet/mods/blob/main/examples.md

Fetch weather data using curl and pipe the JSON output to mods to summarize it into human-readable text, then display with glow. Error output from curl is redirected to /dev/null.

```bash
curl "https://api.open-meteo.com/v1/forecast?latitude=29.00&longitude=-90.00&current_weather=true&hourly=temperature_2m,relativehumidity_2m,windspeed_10m" 2>/dev/null | mods -f "summarize this weather data for a human." | glow
```

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

### Configure Multiple API Providers

Source: https://context7.com/charmbracelet/mods/llms.txt

Define configurations for various LLM API providers in mods.yml, including custom base URLs, API key sources, and model-specific settings like max input characters.

```yaml
# In ~/.config/mods/mods.yml
apis:
  openai:
    base-url: https://api.openai.com/v1
    api-key-env: OPENAI_API_KEY
    # Or use a command to fetch the key
    # api-key-cmd: rbw get openai-api-key
    models:
      gpt-4o:
        aliases: ["4o"]
        max-input-chars: 392000
        fallback: gpt-4
      gpt-4o-mini:
        aliases: ["4o-mini"]
        max-input-chars: 392000

  anthropic:
    base-url: https://api.anthropic.com/v1
    api-key-env: ANTHROPIC_API_KEY
    models:
      claude-sonnet-4-20250514:
        aliases: ["claude-sonnet-4", "sonnet-4"]
        max-input-chars: 680000

  ollama:
    base-url: http://localhost:11434
    models:
      "llama3.2:3b":
        aliases: ["llama3.2"]
        max-input-chars: 650000

  # Azure OpenAI with custom resource
  azure:
    base-url: https://YOUR_RESOURCE.openai.azure.com
    api-key-env: AZURE_OPENAI_KEY
    models:
      gpt-4:
        aliases: ["az4"]
        max-input-chars: 24500

  # OpenAI-compatible endpoints
  custom-endpoint:
    base-url: https://my-llm-proxy.example.com/v1
    api-key: hardcoded-key-if-needed
    models:
      my-model:
        aliases: ["mm"]
        max-input-chars: 50000
```

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

### Override Settings with Environment Variables

Source: https://context7.com/charmbracelet/mods/llms.txt

Control 'mods' behavior by setting environment variables prefixed with MODS_. This allows for dynamic configuration without modifying the config file.

```bash
# Environment variable overrides (prefix with MODS_)
export MODS_MODEL=gpt-4
export MODS_API=anthropic
export MODS_QUIET=true
export MODS_FORMAT=true

# API keys
export OPENAI_API_KEY=sk-xxx
export ANTHROPIC_API_KEY=sk-ant-xxx
export GOOGLE_API_KEY=xxx
export COHERE_API_KEY=xxx
export GROQ_API_KEY=xxx
```

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

### Translate Content

Source: https://context7.com/charmbracelet/mods/llms.txt

Translate text to a different language using Mods with the '--role translator' option.

```bash
echo "Hello, how are you?" | mods --role translator "Translate to Spanish"
```

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

### Set OpenAI API Key

Source: https://github.com/charmbracelet/mods/blob/main/README.md

Set the OPENAI_API_KEY environment variable to authenticate with OpenAI's API. This is required for mods to use GPT-4 or GPT-3.5 Turbo models.

```sh
export OPENAI_API_KEY=your-api-key
```

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

### Code Review with Mods

Source: https://context7.com/charmbracelet/mods/llms.txt

Pipe a git diff to Mods for AI-powered code review and suggestions. The `-f` flag indicates a formatted output is expected.

```bash
git diff HEAD~1 | mods -f "Review this diff for potential issues"
```

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

### Configure HTTP Proxy

Source: https://context7.com/charmbracelet/mods/llms.txt

Specify an HTTP proxy to use for API requests. This is necessary when accessing external services through a proxy server.

```bash
mods --http-proxy "http://proxy.example.com:8080" "My prompt"
```

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

### Key Settings in mods.yml

Source: https://context7.com/charmbracelet/mods/llms.txt

Customize 'mods' behavior by setting various parameters in the mods.yml configuration file, such as default API, model, formatting, and token limits.

```yaml
# Key settings in ~/.config/mods/mods.yml
default-api: openai
default-model: gpt-4o
format: false            # Auto-format as markdown
raw: false               # Disable glamour rendering
quiet: false             # Hide spinner/status
temp: 1.0                # Temperature (0.0-2.0, -1 to disable)
topp: 1.0                # Top-P sampling
max-input-chars: 12250   # Max input characters
max-tokens: 100          # Max response tokens
word-wrap: 80            # Word wrap width
max-retries: 5           # API retry attempts
cache-path: ""           # Custom cache location
```

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

### Show Specific Conversation

Source: https://github.com/charmbracelet/mods/blob/main/features.md

Displays a previous conversation by its title or ID. Exact title match is required, while ID matching uses the first 4 characters.

```bash
mods --show='naturals'
```

```bash
mods -s='a2e2'
```

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

### Control Output Formatting

Source: https://context7.com/charmbracelet/mods/llms.txt

Request formatted output as Markdown or JSON. Use raw mode (`-r`) to disable terminal rendering, which is useful for scripts. The `-p` and `-P` flags can include the prompt or stdin in the output.

```bash
# Request Markdown formatted response (default with -f)
mods -f "Explain the OSI model" | glow
```

```bash
# Request JSON formatted response
mods --format --format-as json "List 3 programming languages with their use cases"
```

```bash
# Raw output without syntax highlighting (useful for scripts)
mods -r "Generate a bash script to backup files"
```

```bash
# Include the prompt in the output
mods -p "What is 2+2?"  # Includes just the args
mods -P 5 "Summarize" < longfile.txt  # Include first 5 lines of stdin
```

```bash
# Quiet mode - suppress spinner and status messages
mods -q "Quick calculation: 15 * 7"
```

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

### Control Top-P (Nucleus Sampling)

Source: https://context7.com/charmbracelet/mods/llms.txt

Fine-tune the LLM's response diversity by setting the Top-P value. This parameter controls the cumulative probability mass of tokens considered for generation.

```bash
# Control Top-P (nucleus sampling) - 0.0 to 1.0
mods --topp 0.9 "Generate diverse ideas"
```

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

### Convert Data Formats

Source: https://context7.com/charmbracelet/mods/llms.txt

Transform data from one format to another (e.g., CSV to JSON) by piping the input file to Mods.

```bash
cat data.csv | mods "Convert this CSV to JSON"
```

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

### Control Top-K Sampling

Source: https://context7.com/charmbracelet/mods/llms.txt

Adjust the 'top-k' parameter to control the randomness of the AI's response. Lower values make the output more focused and deterministic.

```bash
mods --topk 40 "Complete this sentence"
```

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

### Generate Shell Completions

Source: https://github.com/charmbracelet/mods/blob/main/README.md

Generate shell completion scripts for Bash, ZSH, Fish, and PowerShell. These commands help in autocompleting Mods commands in your terminal.

```bash
mods completion bash -h
```

```bash
mods completion zsh -h
```

```bash
mods completion fish -h
```

```bash
mods completion powershell -h
```

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

### Summarize Hacker News Comments

Source: https://github.com/charmbracelet/mods/blob/main/examples.md

Retrieve comments from a Hacker News post using curl and pipe the HTML output to mods to summarize the content, then display with glow. Error output from curl is redirected to /dev/null.

```bash
curl "https://news.ycombinator.com/item?id=30048332" 2>/dev/null | mods -f "what are the authors of these comments saying?" | glow
```

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

### Conversation Branching

Source: https://github.com/charmbracelet/mods/blob/main/features.md

Demonstrates branching conversations by continuing a base conversation with different titles for different output formats.

```bash
mods --title='naturals' 'first 5 natural numbers'
mods --continue='naturals' --title='naturals.json' 'format as json'
mods --continue='naturals' --title='naturals.yaml' 'format as yaml'
```

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

### Analyze Log Files

Source: https://context7.com/charmbracelet/mods/llms.txt

Summarize errors or warnings from log files by piping the relevant lines to Mods for analysis.

```bash
tail -100 /var/log/syslog | mods "Summarize any errors or warnings"
```

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

### List Conversations

Source: https://github.com/charmbracelet/mods/blob/main/features.md

Lists all previous conversations. Can be used with the --list or -l flag.

```bash
mods --list
```

```bash
mods -l
```

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

### Set Custom Title for Mods

Source: https://github.com/charmbracelet/mods/blob/main/features.md

Allows setting a custom title for the current conversation using the --title flag.

```bash
mods --title='title' 'first 2 primes'
```

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

### Set Cohere API Key

Source: https://github.com/charmbracelet/mods/blob/main/README.md

Set the COHERE_API_KEY environment variable to authenticate with Cohere's enterprise-optimized models.

```sh
export COHERE_API_KEY=your-cohere-api-key
```

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

### Set Azure OpenAI Key

Source: https://github.com/charmbracelet/mods/blob/main/README.md

Set the AZURE_OPENAI_KEY environment variable to use Azure OpenAI services. This is an alternative to using the standard OpenAI API key.

```sh
export AZURE_OPENAI_KEY=your-azure-api-key
```

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

### Define Custom Roles in mods.yml

Source: https://context7.com/charmbracelet/mods/llms.txt

Configure custom LLM roles by defining their behavior and constraints in the mods.yml configuration file. These roles can be invoked using the --role flag.

```yaml
roles:
  shell:
    - you are a shell expert
    - you do not explain anything
    - you simply output one liners to solve the problems you're asked
    - you do not provide any explanation whatsoever, ONLY the command

  code-reviewer:
    - you are an expert code reviewer
    - focus on security, performance, and maintainability
    - be constructive and specific in your feedback

  translator:
    - you are a professional translator
    - translate text accurately while preserving tone and context
```

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

### Continue Latest Conversation with New Title

Source: https://github.com/charmbracelet/mods/blob/main/features.md

Continues the latest conversation and saves it with a new title using the --continue=title flag.

```bash
mods 'first 2 primes'
mods --continue='primes as json' 'format as json'
```

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

### Adjust LLM Temperature for Randomness

Source: https://context7.com/charmbracelet/mods/llms.txt

Control the creativity and determinism of the LLM's output by adjusting the temperature parameter. Lower values yield more focused results, while higher values increase randomness.

```bash
# Adjust temperature (randomness) - 0.0 to 2.0
mods --temp 0.2 "Write a formal business email"  # More deterministic
mods --temp 1.5 "Write a creative story"          # More creative
```

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

### Edit Prompt Before Sending

Source: https://context7.com/charmbracelet/mods/llms.txt

Pre-edit your prompt in the default editor before it is sent to the LLM. Useful for complex or sensitive prompts.

```bash
# Edit prompt in your editor before sending
mods -e "Initial prompt"  # Opens $EDITOR
```

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

### Pipe STDOUT from Mods

Source: https://github.com/charmbracelet/mods/blob/main/features.md

Use when STDOUT is not a TTY. The 'Generating' animation goes to STDERR, and the response is streamed to STDOUT.

```bash
echo 'as json' | mods 'first 2 primes' | jq .
```

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

### Use Custom Role in Mods

Source: https://github.com/charmbracelet/mods/blob/main/README.md

This command demonstrates how to use a custom role named 'shell' with the mods tool to perform a specific task, such as listing files.

```sh
mods --role shell list files in the current directory
```

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

### Set Maximum Response Tokens

Source: https://context7.com/charmbracelet/mods/llms.txt

Limit the length of the AI's generated response by setting the 'max-tokens' parameter. This is useful for controlling output size.

```bash
mods --max-tokens 500 "Write a detailed explanation"
```

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

### Pipe STDIN to Mods

Source: https://github.com/charmbracelet/mods/blob/main/features.md

Use when STDIN is not a TTY. The input from STDIN is read and appended to the prompt.

```bash
echo 'as json' | mods 'first 2 primes'
```

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

### Untitled Continue Latest Conversation

Source: https://github.com/charmbracelet/mods/blob/main/features.md

Continues the latest conversation without specifying a new title, useful for quick follow-ups.

```bash
mods 'first 2 primes'
mods --continue-last 'format as json'
```

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

### Configure Retry Attempts

Source: https://context7.com/charmbracelet/mods/llms.txt

Set the maximum number of retries for API requests that might fail. This improves the robustness of automated tasks.

```bash
mods --max-retries 10 "Important request that might fail"
```

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

### Set Stop Sequences

Source: https://context7.com/charmbracelet/mods/llms.txt

Define specific sequences that will cause the AI to stop generating text. This is useful for structured output or preventing run-on sentences.

```bash
mods --stop "END" --stop "###" "Generate until you see a marker"
```

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

### Define a Custom Shell Role

Source: https://github.com/charmbracelet/mods/blob/main/README.md

This YAML configuration defines a custom role named 'shell' for the mods tool. It instructs the AI to act as a shell expert, providing only one-line commands without explanations.

```yaml
roles:
  shell:
    - you are a shell expert
    - you do not explain anything
    - you simply output one liners to solve the problems you're asked
    - you do not provide any explanation whatsoever, ONLY the command
```

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

### Manage Conversations

Source: https://context7.com/charmbracelet/mods/llms.txt

Conversations are automatically saved with SHA-1 identifiers. Continue, branch, show, or delete conversations using title or ID. Old conversations can be deleted based on age.

```bash
# Save with a custom title
mods --title="primes" "What are the first 10 prime numbers?"
```

```bash
# Continue the last conversation
mods -C "Now show them in reverse order"
```

```bash
# Continue a specific conversation by title
mods --continue="primes" "What about the next 5?"
```

```bash
# Continue with a new title (creates a branch)
mods --continue="primes" --title="primes-extended" "Extend to 20 primes"
```

```bash
# List all saved conversations
mods --list
# or
mods -l
```

```bash
# Show a saved conversation by title or ID
mods --show="primes"
mods -s="a2e4"  # First 4+ chars of SHA-1
```

```bash
# Show the last conversation
mods -S
```

```bash
# Delete conversations
mods --delete="primes"
mods --delete="a2e4" --delete="b3f5"
```

```bash
# Delete old conversations
mods --delete-older-than="30d"  # 30 days
mods --delete-older-than="2mo"  # 2 months
```

```bash
# Disable caching for sensitive prompts
mods --no-cache "Analyze this private data" < sensitive.txt
```

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

### Remove Input Character Limit

Source: https://context7.com/charmbracelet/mods/llms.txt

Disable the default input character limit to process very large documents or inputs without truncation.

```bash
mods --no-limit "Process this very long document" < huge-file.txt
```

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

### Generate Commit Messages

Source: https://context7.com/charmbracelet/mods/llms.txt

Use Mods to automatically generate commit messages from staged git changes. This streamlines the version control process.

```bash
git diff --staged | mods "Write a concise commit message for these changes"
```

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

### Delete Conversations

Source: https://github.com/charmbracelet/mods/blob/main/features.md

Deletes one or more conversations by title or ID using the --delete flag. This operation is irreversible.

```bash
mods --delete='naturals' --delete='a2e2'
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.