signalproof-evaluate

Evaluate researched options against explicit criteria, hard gates, and constraints to produce evidence-backed dispositions.

Updated Aug 15, 2026
One-click install
npx skills add https://github.com/docreo/Signalproof-Skills --skill signalproof-evaluate-docreo
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: signalproof-evaluate
Source: https://github.com/docreo/Signalproof-Skills/tree/main/skills/signalproof-evaluate
Command: npx skills add https://github.com/docreo/Signalproof-Skills --skill signalproof-evaluate-docreo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams often pick tools or approaches based on popularity, benchmark winners, or gut feel, hiding uncertainty inside scores and letting licensing, security, or platform blockers get averaged away. This Skill turns gathered research into a defensible decision by enforcing explicit criteria, hard gates, and visible tradeoffs. ## Core Features & Use Cases - Structured Decision Framework: Define the decision, candidates, hard gates, and weighted criteria before scoring, with UNKNOWN preserved instead of fabricated numbers. - Disposition Output: Returns ADOPT, ADAPT, INTEGRATE, PILOT/ISOLATE, DEFER, RESEARCH MORE, REJECT, BUILD FROM SCRATCH, or WATCHLIST with rationale, confidence, and unresolved gates. - Risk-Aware Comparison: Evaluates reversibility, exit cost, readiness burden, security, provenance, and licensing as first-class decision factors. - Use Case: After researching three vector databases, use this Skill to compare them against your platform constraints and licensing requirements, then receive a disposition with confidence level and the smallest next action. ## Quick Start Ask the AI to evaluate the researched candidates against your decision criteria and constraints and recommend a disposition with confidence and unresolved gates.

Frequently Asked Questions about signalproof-evaluate

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

FAQPage Schema
How do I evaluate technology options against decision criteria?▼

Define the decision question and candidates first, then set hard gates that cannot be averaged away, then apply weighted criteria only to surviving options. Score each option on an explicit scale, preserve UNKNOWN for missing evidence, and return a disposition with confidence and unresolved gates.

What is the difference between research and evaluation in decision making?▼

Research establishes what the available evidence supports, while evaluation determines what that evidence means for a specific decision given your criteria and constraints. Research should not secretly pick a winner, and evaluation should not invent missing research.

How do hard gates differ from weighted scoring criteria?▼

Hard gates are non-negotiable blockers such as incompatible licenses, unsupported platforms, or unresolved security issues that cannot be compensated by high scores elsewhere. Weighted criteria express preferences and tradeoffs applied only after all hard gates pass.

Can an ADOPT recommendation authorize installation or purchase?▼

No. A disposition such as ADOPT, ADAPT, or INTEGRATE is a recommendation only and does not authorize downloading, installation, spending, migration, deployment, or any execution. Those actions require separate authority and handoff to the appropriate implementation skill.

When should evaluation return RESEARCH MORE instead of a decision?▼

Return RESEARCH MORE whenever decision-critical evidence is missing, candidate identity is ambiguous, or a hard gate cannot be assessed. Manufacturing certainty from insufficient evidence is treated as a failure condition.