Excellence Loop
Applies a rigorous 10-point quality rubric to every deliverable artifact, identifying gaps and iterating until all criteria score above threshold — transforming good work into excellent work. [EXPLICIT]
Guiding Principle
"Excellence is not a destination but a loop. Ship when every dimension scores above threshold, not when you run out of time."
Procedure
Step 1 — Select the Rubric
- Load the 10-point excellence rubric appropriate for the artifact type. [EXPLICIT]
- For technical documents: completeness, accuracy, evidence, structure, actionability, clarity, diagrams, consistency, audience-fit, next-steps. [EXPLICIT]
- For code deliverables: correctness, readability, testability, performance, security, maintainability, documentation, error-handling, naming, standards-compliance. [EXPLICIT]
- Customize rubric weights based on artifact priority (critical vs. supporting). [EXPLICIT]
- Set the threshold: minimum score per criterion (typically 7/10) and overall (typically 8/10). [EXPLICIT]
Step 2 — Score the Artifact
- Evaluate each criterion independently on a 1-10 scale. [EXPLICIT]
- Provide specific evidence for each score: what is good, what is lacking. [EXPLICIT]
- Identify the lowest-scoring criteria as priority improvement targets. [EXPLICIT]
- Calculate the overall weighted score. [EXPLICIT]
- Determine the verdict: PASS (all criteria >= threshold), ITERATE (specific criteria below), REWORK (overall below threshold). [EXPLICIT]
Step 3 — Iterate on Gaps
- For each criterion below threshold, define the specific improvement needed. [EXPLICIT]
- Apply improvements in priority order (lowest scores first). [EXPLICIT]
- Re-score only the modified criteria after each improvement pass. [EXPLICIT]
- Track iteration count — if more than 3 iterations on the same criterion, escalate for review. [EXPLICIT]
- Document what changed in each iteration for audit trail. [EXPLICIT]
Step 4 — Certify and Archive
- When all criteria pass, stamp the artifact with the excellence certification. [EXPLICIT]
- Record the final scores and iteration history. [EXPLICIT]
- Add the ghost menu with next-step actions. [EXPLICIT]
- Update the artifact status from {WIP} to ready for review or {Aprobado}. [EXPLICIT]
- Feed the scoring data into the lessons-learned tracker for continuous improvement. [EXPLICIT]
Quality Criteria
- Every artifact is scored against a defined rubric before delivery.
- No criterion scores below the minimum threshold in the final version.
- Iteration history is documented showing progressive improvement.
- Rubric is calibrated quarterly based on feedback and lessons learned.
Anti-Patterns
- Skipping the excellence loop due to time pressure ("good enough" mentality).
- Scoring all criteria at 10/10 without honest assessment (rubber stamping).
- Iterating endlessly on minor improvements (diminishing returns).
- Using the same rubric for all artifact types without customization.
Usage
Example invocations:
- "/excellence-loop" — Run the full excellence loop workflow
- "excellence loop on this project" — Apply to current context
Assumptions & Limits
- Assumes access to project artifacts (code, docs, configs) [EXPLICIT]
- Requires English-language output unless otherwise specified [EXPLICIT]
- Does not replace domain expert judgment for final decisions [EXPLICIT]
Edge Cases
| Scenario | Handling |
|---|
| Empty or minimal input | Request clarification before proceeding |
| Conflicting requirements | Flag conflicts explicitly, propose resolution |
| Out-of-scope request | Redirect to appropriate skill or escalate |