never-fabricate
CommunityVerify before you report.
AuthorEtanHey
Version1.0.0
Installs0
System Documentation
What problem does it solve?
This Skill prevents the AI from making false claims about file contents, test results, or agent outputs by enforcing a strict "read-then-report" protocol.
Core Features & Use Cases
- Grounded Reporting: Ensures all statements about files, tests, or agent work are based on actual inspection, not assumptions.
- Prevents Fabrication: Stops the AI from claiming "all green" or "complete" without verifying the evidence.
- Use Case: When asked if a code branch is "green" after tests, this Skill forces the AI to read the test output file and report accurately, even if a test failed.
Quick Start
Use the never-fabricate skill to confirm the status of the latest test run by reading the provided output file.
Dependency Matrix
Required Modules
None requiredComponents
references
💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: never-fabricate Download link: https://github.com/EtanHey/golems/archive/main.zip#never-fabricate Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.