Token Budget Advisor for Openclaw

An AI agent utility that intercepts responses to let users choose their preferred level of detail and token consumption.

djc00p
v1.0.1
Apr 6, 2026
0
937
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install token-budget-advisor

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install token-budget-advisor using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Token Budget Advisor?

Token Budget Advisor is a strategic utility for AI coding agents that allows for precise control over output volume. By prompting the user to select a response depth before generating content, it ensures that token usage aligns with the user's immediate needs. This skill, part of the Openclaw Skills ecosystem, helps prevent the unnecessary cost and latency associated with overly verbose AI responses when a simple summary would suffice.

Built on a heuristic estimation model, it provides an approximate token count for different response tiers ranging from essential summaries to exhaustive deep dives. It is an ideal addition to any developer's Openclaw Skills toolkit for maintaining efficiency during long-form coding sessions or research tasks.

Token Budget Advisor Use Cases

  • Controlling costs by limiting token consumption on routine queries.
  • Forcing a quick TL;DR or summary for complex technical documentation.
  • Requesting exhaustive, multi-example analysis for critical architectural changes.
  • Managing session-wide response length preferences without repeating instructions.

How Token Budget Advisor Works

  1. The skill monitors incoming prompts for keywords like budget, depth, or response length.
  2. It analyzes the input complexity to estimate the potential response size using an 85-90% accurate heuristic.
  3. It intercepts the response flow and presents the user with four depth options: Essential (25%), Moderate (50%), Detailed (75%), and Exhaustive (100%).
  4. Upon user selection or the use of a shortcut, the skill instructs the agent to format the output according to the chosen constraints.
  5. The selected depth level is persisted for the duration of the session to maintain consistency.

Token Budget Advisor Setup

To add this capability to your environment, install it via the clawdbot CLI within the Openclaw Skills framework:

clawdbot install token-budget-advisor

No complex configuration or API keys are required for this skill to function.

Token Budget Advisor Data Schema & Taxonomy

The Token Budget Advisor organizes its response logic based on the following taxonomy:

Level Depth % Content Includes Target Length
Essential 25% Direct answer and conclusion only 2-4 sentences
Moderate 50% Answer, context, and one example 1-3 paragraphs
Detailed 75% Examples, pros/cons, and alternatives Structured list
Exhaustive 100% Full analysis, edge cases, and all context No limit

Token Budget Advisor Advanced Features

  • Shortcut recognition for immediate depth setting using terms like brief, tldr, or numeric values.
  • Session-based state persistence to remember user depth preferences.
  • Heuristic-based token estimation that works without requiring a local tokenizer.
  • Broad compatibility across Linux, macOS, and Windows environments within Openclaw Skills.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*