skills.rest
w-pinkietech
w-pinkietechOFFICIAL
Explore all skills by this author →
GitHub

sanitize-artifacts

Removes prompt leakage and production residue from generated artifacts.

1|Updated Mar 25, 2026
Search Tags:#editing#content cleanup#artifact#deliverable#revision#sanitize#prompt leakage
One-click install
npx skills add https://github.com/w-pinkietech/iotkit --skill sanitize-artifacts-w-pinkietech
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: sanitize-artifacts
Source: https://github.com/w-pinkietech/iotkit/tree/main/.agents/skills/sanitize-artifacts
Command: npx skills add https://github.com/w-pinkietech/iotkit --skill sanitize-artifacts-w-pinkietech

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Artifacts produced through iterative prompting often leak conversation history, prompt instructions, constraints, and corrective feedback into the final deliverable, making it read like a patchwork of the production process rather than a coherent standalone document. ## Core Features & Use Cases - Leakage Detection: Identifies phrases that expose prompts, constraints, avoided approaches, tool choices, and meta-commentary that the intended audience never needs. - Intent-Based Rewriting: Converts production guidance (constraints, examples, feedback) into natural artifact design through tone, structure, and wording rather than explicit disclaimers. - Use Case: After drafting a user guide through several rounds of corrective prompting, run this Skill to strip out sentences like "As requested, this guide avoids CLI tools" and rewrite them as direct, audience-facing instructions. ## Quick Start Ask the assistant to sanitize the artifact just produced in this session so it reads as a clean standalone deliverable.

Frequently Asked Questions about sanitize-artifacts

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

FAQPage Schema
How do I remove prompt leakage from AI-generated documents?▼

Review the artifact for phrases referencing instructions, constraints, or conversation history, then rewrite them as direct audience-facing content. This Skill classifies each statement as user-facing content, invisible guidance, or residue and edits accordingly.

What counts as production residue in a generated artifact?▼

Production residue includes phrases like "As requested", "This avoids", "The user wanted", explicit constraint disclaimers, and examples copied from the conversation. These reveal how the artifact was made rather than serving its audience.

Should constraints ever appear explicitly in the final artifact?▼

Only when the intended audience genuinely needs to know them. Otherwise constraints should shape the artifact's design, tone, and wording invisibly, such as writing Drive-based steps instead of stating that Git is not used.

When should I not use artifact sanitization?▼

Do not sanitize when the meta-information is genuinely part of the deliverable, such as a change log, a process report the user requested, or documentation that must explain design decisions to its readers.

Does sanitizing an artifact change its technical content?▼

No. The revision preserves the artifact's intended purpose, technical correctness, and necessary user-facing requirements. Only production residue and meta-commentary are removed or rewritten into natural wording.

Related Skills

Explore More Content & Communication Skills→
HyperionBurnHyperionBurn

ralph

Run PRD-driven persistence loops until all acceptance criteria pass reviewer verification.

Community
Advanced
grafanagrafana
5.5k

persona-check

Assess documentation content for fit against target audience personas.

Official
Intermediate
jikig-aijikig-ai
12

every-style-editor

Performs line-by-line editing with automatic detection of Every style-rule violations.

Official
Intermediate
open-gsdopen-gsd
1.0k

create-workflow

Draft valid YAML workflow definitions from user prompts.

Official
Intermediate