skills.rest
DawnMoon1542
DawnMoon1542COMMUNITY
Explore all skills by this author →
GitHub

full-output-enforcement

Enforces complete, unabridged LLM output by banning placeholder patterns and truncation.

7|Updated May 13, 2026
Search Tags:#code generation#token limit#llm output#placeholder detection#anti-truncation#complete output
One-click install
npx skills add https://github.com/DawnMoon1542/agents-skills --skill full-output-enforcement-dawnmoon1542
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: full-output-enforcement
Source: https://github.com/DawnMoon1542/agents-skills/tree/main/full-output-enforcement
Command: npx skills add https://github.com/DawnMoon1542/agents-skills --skill full-output-enforcement-dawnmoon1542

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? LLMs often truncate long responses, insert placeholder comments like "// rest of code", or skip deliverables to save tokens. This Skill overrides that default behavior so every requested file, function, or section is delivered in full. ## Core Features & Use Cases - Banned Pattern Enforcement: Blocks placeholder comments, ellipsis shortcuts, and prose like "for brevity" or "the rest follows the same pattern". - Scope Locking and Cross-Check: Counts expected deliverables before generating and verifies the count before responding. - Clean Token-Limit Handling: Pauses at a clean breakpoint with a resume marker instead of compressing or skipping content. - Use Case: Ask for a full multi-file project scaffold or five complete React components, and receive every item fully implemented with no skeletons or omissions. ## Quick Start Apply the full-output-enforcement skill and generate the complete implementation of all requested files with no placeholders or omissions.

Frequently Asked Questions about full-output-enforcement

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

FAQPage Schema
How do I stop an LLM from truncating code output?▼

Use explicit anti-truncation instructions that ban placeholder patterns like "// rest of code" and require every requested deliverable in full. This Skill locks the deliverable count before generating and cross-checks it before responding.

How to handle LLM responses that hit the token limit?▼

Write at full quality up to a clean breakpoint such as the end of a function or file, then emit a pause marker stating progress and the next section. On "continue", resume exactly where you stopped without recap or repetition.

What placeholder patterns should be banned in AI code generation?▼

Ban comments like "// ...", "// TODO", "// implement here", and "// similar to above", plus prose like "for brevity" and "the rest follows the same pattern". These patterns indicate omitted content that breaks runnable output.

When should I not use full-output enforcement?▼

Avoid it for exploratory questions, quick prototypes, or when a concise summary is genuinely preferred. Forcing exhaustive output on simple questions wastes tokens and slows iteration.

Related Skills

Explore More Software Engineering Skills→
jshsakurajshsakura
22

business-analyst

Clarify requirements, normalize scope, and define acceptance criteria from messy inputs.

Community
Intermediate
HKUST-QUANT-SOCIETYHKUST-QUANT-SOCIETY

compose:verify

Enforces running verification commands before claiming work is complete or passing.

Official
Basic
agno-agiagno-agi
41.6k

code-review

Review Python and multi-language code for style, bugs, and security concerns.

Official
Intermediate
githubgithub
38.5k

first-ask

Refines task scope through interactive human input before executing work.

Official
Basic