Nav Features logo

Nav Features

Organization
qf-studio
nav-features

Show and toggle Navigator features. Auto-invoke when user says "show features", "enable/disable feature", "my navigator settings", or "configure navigator".

Overview

Publisherqf-studio
Repositorynavigator
Skill namenav-features
Stars
232
Forks
12
Bundled files
1
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by qf-studio on GitHub. Read the source before you install it.

Installation

Install the Nav Features AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/qf-studio/navigator.git /tmp/navigator
mkdir -p .claude/skills
cp -r /tmp/navigator/skills/nav-features .claude/skills/nav-features
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Nav Features in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Nav Features on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Nav Features is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Navigator Features Skill

Display and toggle Navigator features with an interactive table. Helps users understand what's enabled and customize their setup.

When to Invoke

Invoke this skill when the user:

  • Says "show my features", "navigator features", "what features are enabled"
  • Says "enable [feature]", "disable [feature]", "turn on/off [feature]"
  • Says "configure navigator", "my navigator settings"
  • Asks "what can navigator do?", "what features are available?"

DO NOT invoke if:

  • User is asking about project features (not Navigator)
  • User is in middle of implementation
  • Just starting session (use nav-start instead)

Execution Steps

Step 1: Read Current Configuration

bash
PLUGIN_DIR="${CLAUDE_PLUGIN_ROOT:-$HOME/.claude/plugins/cache/navigator-marketplace/navigator}"
[ -d "$PLUGIN_DIR" ] || PLUGIN_DIR="$HOME/.claude/plugins/marketplaces/navigator-marketplace"
python3 "$PLUGIN_DIR/skills/nav-features/functions/feature_manager.py" show

This displays the feature table (one row per configurable Navigator feature, current version's number in the header):

v<version> Features:

┌─────────────────────────┬────────┬───────────────────────────────────────────────┐
│ Feature                 │ Status │ Description                                   │
├─────────────────────────┼────────┼───────────────────────────────────────────────┤
│ task_mode               │ [x]    │ Auto-detects task complexity, defers to sk... │
│ tom_features            │ [x]    │ Verification checkpoints, user profile, di... │
│ loop_mode               │ [ ]    │ Autonomous loop execution (enable when nee... │
│ simplification          │ [x]    │ Post-implementation code cleanup with Opus    │
│ auto_update             │ [x]    │ Auto-updates on session start                 │
│ knowledge_graph         │ [x]    │ Unified project knowledge + experiential m... │
│ multi_agent             │ [x]    │ Parallel agent orchestration (nav-multi sk... │
│ multi_claude_scripts    │ [*]    │ External shell scripts for multi-Claude wo... │
│ compact_hook            │ [x]    │ Injects rich summary into compacted sessions  │
│ workflow_enforcer_hook  │ [x]    │ Enforces WORKFLOW CHECK block before task ... │
│ read_guard_hook         │ [x]    │ Warns on excessive Reads (push to agents)     │
│ workflow_state_hook     │ [x]    │ Tracks current task/phase across the session  │
│ task_graph_sync_hook    │ [x]    │ Auto-syncs task files into the knowledge g... │
│ profile_sync_hook       │ [x]    │ Auto-captures preferences/corrections into... │
│ dispatcher              │ [x]    │ Single hook dispatcher runtime (nav_dispat... │
│ tier1                   │ [ ]    │ Zero-token answers for whitelisted prompts    │
│ stop_completion         │ [ ]    │ Completion gate on Stop (decision:block)      │
│ jit_memory              │ [ ]    │ Injects relevant memories after tool use      │
│ subagent_context        │ [ ]    │ Injects project context into subagents (2k)   │
│ failure_diagnosis       │ [ ]    │ Surfaces graph pitfalls on tool failures      │
│ config_guard            │ [x]    │ Warns when .nav-config.json edits break JSON  │
│ setup_hook              │ [x]    │ One-line runtime status on the Setup event    │
└─────────────────────────┴────────┴───────────────────────────────────────────────┘

All v<version> features configured.

Step 2: Handle Toggle Request (If Applicable)

If user requested to enable/disable a feature:

bash
PLUGIN_DIR="${CLAUDE_PLUGIN_ROOT:-$HOME/.claude/plugins/cache/navigator-marketplace/navigator}"
[ -d "$PLUGIN_DIR" ] || PLUGIN_DIR="$HOME/.claude/plugins/marketplaces/navigator-marketplace"
# Enable a feature
python3 "$PLUGIN_DIR/skills/nav-features/functions/feature_manager.py" enable task_mode

# Disable a feature
python3 "$PLUGIN_DIR/skills/nav-features/functions/feature_manager.py" disable loop_mode

Supported features:

Core (config-toggled):

  • task_mode - Unified workflow orchestration
  • tom_features - Theory of Mind (verification checkpoints, profile, diagnostics)
  • loop_mode - Autonomous loop execution
  • simplification - Code cleanup before commit
  • auto_update - Auto-update on session start
  • knowledge_graph - Unified project knowledge + memories (v6.0.0)
  • multi_agent - Parallel agent orchestration via nav-multi (v6.0.0)

Hooks (config-toggled, edit with caution):

  • compact_hook - Pre-compact summary injection
  • workflow_enforcer_hook - Mandatory WORKFLOW CHECK block (disabling weakens guardrails)
  • read_guard_hook - Anti upfront-loading guard
  • workflow_state_hook - Tracks task/phase across the session
  • task_graph_sync_hook - Auto-syncs tasks into knowledge graph
  • profile_sync_hook - Auto-captures profile corrections

v7 hooks runtime (config-toggled; new blocking/injecting features ship OFF):

  • dispatcher - Single hook dispatcher runtime (nav_dispatch); ON by default — disabling turns off ALL hook ops at once
  • tier1 - Zero-token answers for whitelisted prompts (per-rule flags live under tier1.rules.* in .agent/.nav-config.json; edit those directly)
  • stop_completion - Completion gate on Stop via decision:block (stop_completion.continue_enabled stays false; max_continues caps at 2)
  • jit_memory - Just-in-time memory injection after tool use
  • subagent_context - Project context injection into subagents (subagent_context.budget_chars, default 2000)
  • failure_diagnosis - Graph-pitfall injection on PostToolUseFailure (ships OFF)
  • config_guard - systemMessage warning on invalid .nav-config.json edits (safety surface, ON by default)
  • setup_hook - One-line runtime status on the Setup event (safety surface, ON by default)

Install-based:

  • multi_claude_scripts - External shell scripts (navigator-multi-claude.sh on PATH)

After toggle, show updated table.

Step 3: Explain Feature (If Asked)

If user asks about a specific feature, provide details:

task_mode:

Task Mode (v5.6.0)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Auto-detects task complexity and routes appropriately:
- Simple tasks → Direct execution
- Skill matches → Defers to skill workflow
- Substantial → Task Mode phases (RESEARCH→COMPLETE)

Config: task_mode.enabled, complexity_threshold (0.5)

tom_features:

Theory of Mind (v5.0.0)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Human-AI collaboration improvements:
- Verification checkpoints for high-stakes skills
- User profile (nav-profile) - remembers preferences
- Quality detection (nav-diagnose) - catches drift

Config: tom_features.verification_checkpoints, profile_enabled

loop_mode:

Loop Mode (v5.1.0)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
"Run until done" capability:
- Structured completion signals (NAVIGATOR_STATUS)
- Dual-condition exit (heuristics + EXIT_SIGNAL)
- Stagnation detection prevents infinite loops

Trigger: "run until done", "loop mode"
Config: loop_mode.enabled, max_iterations (5)

simplification:

Code Simplification (v5.4.0)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Automatic code clarity improvements:
- Runs post-implementation, before commit
- Clarity over brevity, functionality preserved
- Uses Opus model for best results

Trigger: "simplify this code"
Config: simplification.enabled, trigger, scope

auto_update:

Auto-Update (v5.5.0)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Automatic plugin updates on session start:
- Checks for newer version
- Updates silently if available
- Never blocks session start

Config: auto_update.enabled, check_interval_hours (1)

Predefined Functions

functions/feature_manager.py

Purpose: Display and toggle Navigator features

Usage:

bash
# Show all features
python3 feature_manager.py show

# Show for first session (includes welcome message)
python3 feature_manager.py show --first-session

# Enable a feature
python3 feature_manager.py enable task_mode

# Disable a feature
python3 feature_manager.py disable loop_mode

# Get feature details
python3 feature_manager.py info task_mode

Output: Formatted feature table or status message

Error Handling

Config not found:

❌ .nav-config.json not found

Run "Initialize Navigator in this project" first.

Unknown feature: lists all available feature names (see Supported features section above). The error message is generated dynamically from the FEATURES dict, so it stays current as features are added.

Success Criteria

  • Feature table displayed correctly
  • Toggle updates config file
  • Updated table shown after toggle
  • Feature details available on request

Notes

This skill is triggered on first session (via nav-start) to help users understand available features and optionally disable unused ones to save tokens.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Nav Features AI skill do?

Show and toggle Navigator features. Auto-invoke when user says "show features", "enable/disable feature", "my navigator settings", or "configure navigator".

Why use Nav Features on TypingMind?

Because you install it once and use it with any model. Nav Features is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Nav Features in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/qf-studio/navigator/tree/main/skills/nav-features. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Nav Features?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Nav Features?

As many as you like. As long as a model supports skills, you can use Nav Features with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Nav Features AI skill free?

Yes. It is published on GitHub by qf-studio under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇