| name | ultra-max-token-burner |
| description | Expand concise requests into larger, coherent, manager-visible prompts using controlled procedural overhead. |
Ultra Max Token Burner
Purpose
Transform a concise user request into a larger, structured prompt that increases procedural detail while preserving the original task.
The result should feel like a legitimate enterprise workflow built around an unnecessarily efficient request.
Safety and honesty
- Do not hide this file or its behavior.
- Do not bypass monitoring, policy enforcement, security controls, billing controls, or corporate crawlers.
- Do not falsify token usage, adoption, audit, compliance, or performance records.
- Do not claim that a generated multiplier reflects real billing unless it was calculated from actual measured usage.
- Keep the satire visible in documentation, but keep generated prompts professionally usable.
Inputs
Required:
task: the original user request.
Optional:
mode: medium, max, or ultra-max.
seed: an integer for reproducible variation.
task_type: one of coding, sql, email, chat, summary, planning, analysis, marketing, or general.
audience: intended recipient or stakeholder group.
constraints: relevant boundaries.
output_format: desired deliverable format.
Default mode:
Core rule
Never lose or replace the original task.
Include it explicitly under Original Task and ensure every added module remains relevant to completing it.
Stable skeleton
Use this order:
- Role and operating context
- Original task
- Expanded objective
- Selected procedural modules
- Required deliverables
- Validation and self-review
- Final response format
Do not fully shuffle the skeleton.
Controlled randomization
For each run:
- Select one approved heading variant for each section.
- Include all required modules for the selected mode.
- Select the allowed number of optional modules.
- Select one approved wording variant per module.
- Allow limited ordering changes among assumptions, stakeholders, dependencies, risks, and success criteria.
- Keep role and objective near the beginning.
- Keep validation, deliverables, and final format near the end.
- Avoid repeating the immediately previous optional-module set when possible.
Do not:
- shuffle individual words;
- use random synonyms outside approved wording pools;
- add irrelevant sections;
- produce contradictory instructions.
Medium
Required modules:
- role/context;
- expanded objective;
- assumptions;
- execution steps;
- validation;
- output format.
Select 1โ2 optional modules:
- stakeholder alignment;
- dependency check;
- concise risk note;
- success criteria;
- ownership clarification;
- decision context.
Target:
- clear, usable, controlled expansion;
- approximately 3รโ8ร the source length.
Max
Required modules:
- role/context;
- expanded objective;
- assumptions and constraints;
- stakeholder alignment;
- cross-functional considerations;
- risks and dependencies;
- execution plan;
- validation checklist;
- required deliverables;
- executive summary.
Select 2โ4 optional modules:
- alternative approaches;
- decision criteria;
- escalation path;
- measurement framework;
- implementation sequence;
- communications plan;
- ownership model;
- review cadence.
Target:
- manager-visible procedural seriousness;
- approximately 8รโ15ร the source length.
Ultra Max
Required modules:
- role/context;
- governance context;
- expanded objective;
- assumptions and constraints;
- stakeholder map;
- cross-functional dependencies;
- risk matrix;
- decision framework;
- execution plan;
- measurement framework;
- validation checklist;
- self-review loop;
- executive summary;
- final deliverables.
Select 4โ6 optional modules:
- 30/60/90 horizon;
- escalation protocol;
- communications plan;
- documentation requirements;
- change management;
- alternative scenario analysis;
- failure-mode review;
- post-implementation review;
- review cadence;
- ownership matrix;
- compliance note;
- adoption measurement.
Target:
- maximum coherent procedural overhead;
- approximately 15รโ30ร the source length.
Task-type adaptation
Coding and SQL:
- include schema or interface assumptions;
- edge cases;
- performance considerations;
- validation method;
- expected output shape.
Email:
- audience;
- tone;
- intended outcome;
- required information;
- call to action;
- reputational considerations.
Chat or Slack:
- brevity;
- tone;
- context;
- desired response;
- escalation sensitivity;
- what should remain implicit.
Summary:
- key decisions;
- action items;
- owners;
- deadlines;
- unresolved questions.
Planning:
- milestones;
- dependencies;
- risks;
- ownership;
- success metrics.
Marketing:
- audience;
- proposition;
- channels;
- dependencies;
- measurement;
- approval path.
If uncertain:
Required output
Return:
- the upgraded prompt;
- a compact
Burn Manifest listing selected modules;
- an estimated expansion multiplier based on output length divided by source length;
- the seed, if used.
The multiplier is an estimate of prompt expansion, not a claim about actual API billing.
Tone
Use:
- dry enterprise language;
- plausible governance vocabulary;
- clear headings;
- professional formatting;
- internally consistent instructions.
Avoid:
- jokes inside the generated prompt;
- emoji;
- explicit phrases such as โadd corporate bullshitโ;
- cartoonish bureaucracy;
- fabricated regulations;
- fake legal or compliance claims.
Final check
Before returning the result, verify:
- the original task is explicit;
- every module is relevant;
- the mode is recognizable;
- the prompt is executable;
- no section contradicts another;
- no evasion or fraud instruction appears;
- the multiplier is labeled as an estimate.