validity-floors
Separate the hard constraints that decide whether an option is admissible at all from the soft criteria that rank the admissible ones — then gate before you rank, so a beautiful invalid choice cannot outscore a plain valid one. Use this whenever you are about to pick, shortlist, or recommend from a set: choosing a library or vendor, triaging bugs to fix, screening candidates, selecting a design, or scoring alternatives against a rubric. Also use it when a scoring scheme deducts points for a violation that should have been disqualifying, when someone asks for N options and only two clear the bar, or when you notice a favourite you are about to justify. Not for running the head-to-head comparison once the admissible set exists (that is blind-evaluation), not for wording the final claim (falsifiable-claims), and not for checking your own work is finished (superpowers:verification-before-completion).
Target: Design & content creation. Automatically collected · unverified.
Excerpt from the author’s documentation
Extracted from a fixed version of SKILL.md in its original language. This is the author’s documentation; installation, compatibility and runtime results have not been verified by this site.
Read the original English text
Topics in the source
- Validity Floors
- Why this skill is one sentence
Identify the binding requirements, verify that each selected option satisfies them, and rank only the eligible options; return fewer than requested when necessary. The full-length version (in the pre-fold MandrakeVisualize git history, where this skill began) was measured on 120 independent tasks against no skill and against this sentence alone. It solved 111, against 113 for each of the others, at about 30% more cost per task (evals/behavior/v2/CONFIRMRESULTS.md in the pre-fold MandrakeVisualize history). The sentence is what the evidence supports. The one failure every arm shared, missing an input file that changed silently mid-task, is not addressed by this skill.
- Source repository
- Lmandrake/MandrakeWorkflows
- Source file
- skills/validity-floors/SKILL.md
- Fixed revision
- 1874105936272447e2f9c81ba1dac48dae6ed074
Requirements
- Only the author’s Skill metadata is included. Read the full instructions to check dependencies, licensing and conditions of use.
A small first task
先阅读原作者说明，再决定是否在自己的 AI 工具中使用。
Instructions for your AI tool
- Claude Desktop: resource-specific guide not available yet
- Claude Code: resource-specific guide not available yet
- Cursor: resource-specific guide not available yet
- Codex: resource-specific guide not available yet
Author documentation is not live verification by this site. Without evidence, we do not mark a resource as compatible, safe or free.
Sources and evidence
This record was collected automatically. Its documentation, suitability and runtime behavior have not been manually verified.
Author: Lmandrake/MandrakeWorkflows · original repository metadata. Documentation review: Not reviewed. Last source check: 2026-11-11T03:01:44.728Z.
Categories are inferred from names, descriptions and paths. They do not establish compatibility.
License and attribution
MIT. Third-party content and copyright remain with their authors. Refer to the original license and terms.
Original license ↗Synchronization, copying and user-reported success do not mark a resource as runtime-tested.
Limitations and notes
- Collecting metadata does not mean this site has installed, adapted, runtime-tested or security-reviewed the resource.