pql-framework

I've been looking at your pql-framework skill, and I'm curious about the decision to abstract query language primitives the way you have—most frameworks I see either go full DSL or stick with string building, but you're threading something in between.

Links:

• GitHub
• Your agentic skill in the SkillzWave marketplace

The TL;DR

You're at 65/100, which puts you in D territory. This skill addresses a real PLG problem—defining and scoring product-qualified leads—but it's currently too conceptual. Your strongest area is Writing Style (8/10), which means the prose is clean and professional. The weakest area by far is Utility (11/20)—templates are mentioned but not provided, and there are no concrete examples or validation patterns developers can actually implement.

What's Working Well

• Clear conceptual structure. The 5-step framework (Signal Library → Scoring Model → Routing Logic → Integration → Measurement Loop) is sound and easy to follow.
• Consistent terminology. PQL/PQA language is used cleanly throughout without unnecessary jargon.
• Professional tone. The writing is straightforward and objective—no marketing fluff, which is refreshing.

The Big One: Missing Templates and Examples

This is the issue holding you back the most. You mention "Signal inventory worksheet," "Scoring matrix with weights and decay logic," and "Routing decision tree," but none of these templates actually exist. A developer reading your skill knows conceptually what a PQL program should include, but they can't execute it without seeing actual templates.

Fix: Create three reference files:

• references/signal-inventory-template.md – Sample signals with point values (e.g., "Invited 3+ users (5 pts), Hit workspace limit (8 pts)")
• references/scoring-matrix-template.md – Actual scoring table with decay rules and thresholds
• references/routing-decision-tree.md – Decision logic linking tiers to specific plays

This alone would add 7 points and make the skill actually useful.

Other Things Worth Fixing

1. Add trigger phrases to your description. Right now it's "Methodology for defining product-qualified lead signals..." Users search for "score leads," "PQL program," "qualify accounts." Update the description to include these terms for discoverability (+2 points).

2. No validation checklist. The Measurement Loop step mentions tracking conversion but doesn't explain how to validate your scoring model actually works. Add a references/validation-checklist.md with steps like: run on historical cohort → compare against actual conversions → identify false positives → adjust weights (+3 points).

3. Vague sub-steps in the framework. "Weight signals, set decay rules" is clear at 50,000 feet but lacks the decision points someone new to PQL would need (e.g., "1-10 point scale?" "30-day or event-based decay?"). Add a bit more tactical detail here.

Quick Wins

• Add three reference template files (biggest bang for buck)
• Strengthen the description with specific trigger phrases
• Create a validation checklist so users know how to verify their implementation
• Expand framework steps with concrete decision points

Checkout your skill here: SkillzWave.ai | SpillWave We have an agentic skill installer that install skills in 14+ coding agent platforms. Check out this guide on how to improve your agentic skills.