skills.rest
riftzen-bit
riftzen-bitCOMMUNITY
Explore all skills by this author →
GitHub

token-budget-advisor

Estimate token needs and enforce selectable response depth levels.

1|Updated Apr 7, 2026
Search Tags:#tokens#cost-optimization#token-budget#depth-control#response-length#heuristic-estimation#session-preference
One-click install
npx skills add https://github.com/riftzen-bit/gemini-setup --skill token-budget-advisor-riftzen-bit
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: token-budget-advisor
Source: https://github.com/riftzen-bit/gemini-setup/tree/main/skills/token-budget-advisor
Command: npx skills add https://github.com/riftzen-bit/gemini-setup --skill token-budget-advisor-riftzen-bit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Helps users choose how much of the model's output to consume by estimating prompt and response token needs and offering selectable depth levels before the assistant answers, preventing unexpected long outputs and conserving token budget.

Core Features & Use Cases

  • Heuristic Token Estimation: Estimates input tokens using prose and code heuristics and maps to a response window using configurable multipliers.
  • User Depth Selection: Presents 4 depth options (Essential/Moderate/Detailed/Exhaustive) with approximate token targets and accepts direct shortcuts like "25%" or "tldr".
  • Session Persistence & Multilingual Triggers: Remembers chosen depth for the session and recognizes multilingual trigger phrases for on-demand control.
  • Use Case: Ask for a short executive summary to limit cost, or select exhaustive depth for full code reviews or architectural analyses.

Quick Start

Ask the assistant "Respond at 50% depth" to receive a balanced answer with context and one example.

Frequently Asked Questions about token-budget-advisor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I control token consumption and response length during AI conversations?▼

You control token consumption by selecting from four depth levels—Essential, Moderate, Detailed, or Exhaustive—which sets heuristic token targets and enforces that response length for the session.

How does heuristic token estimation work for prompt and response windows?▼

Heuristic token estimation calculates input tokens using simple prose and code heuristics, then maps the result to a response window using configurable multipliers to predict output size.

Can I request a specific response depth using shortcuts or multilingual phrases?▼

Yes, you can use direct shortcuts like "25%" or "tldr" and multilingual trigger phrases to instantly adjust response depth and token limits on-demand during your session.

What is the best way to conserve token budget for executive summaries versus full code reviews?▼

Select Essential or Moderate depth for short executive summaries to minimize cost, and choose Exhaustive depth when you need full outputs for comprehensive code reviews or architectural analyses.

Does the selected response depth persist across multiple prompts in the same session?▼

Yes, session persistence remembers your chosen depth level, automatically applying the same token budget constraints to subsequent responses until you explicitly change the setting.

When should I avoid using fixed token budget limits for AI responses?▼

Avoid fixed token budget limits when tasks require unbounded architectural analyses or exhaustive code generation, as strict depth control may truncate necessary context and critical implementation details.

Related Skills

Explore More Software Engineering Skills→
ElliotJLTElliotJLT
60

dont-be-greedy

Estimate token costs for CSV, JSON, and TXT files before loading.

Community
Intermediate
starrank-softstarrank-soft

token-budget-advisor

Intercept AI response generation to offer selectable depth levels based on estimated token consumption.

Official
Intermediate
acefrost511acefrost511

skill-evolution-manager

Analyze user feedback and session insights to update skill definitions.

Community
Advanced
gganbukim1gganbukim1

token-budget-advisor

Estimate prompt size and map complexity to selectable output-length tiers.

Community
Basic