| name | skill-101 |
| description | Patterns and techniques for evaluating and improving AI agent outputs. Use this skill when implementing self-critique and reflection loops, building evaluator-optimizer pipelines for quality-critical generation, creating test-driven code refinement workflows, designing rubric-based or LLM-as-judge evaluation systems, adding iterative improvement to agent outputs (code, reports, analysis), and measuring agent response quality. |
Agentic Evaluation Patterns
Patterns for self-improvement through iterative evaluation and refinement.
Workspace-baseline context: This skill supports the template quality workflow.
Use alongside @Code Review Agent (quality checklist validation),
@SpecKit Analyze (cross-artifact consistency), and the Quality Checklist
in copilot-instructions.md to evaluate generated templates against baseline standards.
Overview
Evaluation patterns enable agents to assess and improve their own outputs, moving beyond single-shot generation to iterative refinement loops.
Generate → Evaluate → Critique → Refine → Output
↑ │
└──────────────────────────────┘
When to Use
- Quality-critical generation: Code, reports, analysis requiring high accuracy
- Tasks with clear evaluation criteria: Defined success metrics exist
- Content requiring specific standards: Style guides, compliance, formatting
- Template validation: Checking new instructions/agents/prompts/skills against baseline Hard Rules and Quality Checklist
Pattern 1: Basic Reflection
Agent evaluates and improves its own output through self-critique.
def reflect_and_refine(task: str, criteria: list[str], max_iterations: int = 3) -> str:
"""Generate with reflection loop."""
output = llm(f"Complete this task:\n{task}")
for i in range(max_iterations):
critique = llm(f"""
Evaluate this output against criteria: {criteria}
Output: {output}
Rate each: PASS/FAIL with feedback as JSON.
""")
critique_data = json.loads(critique)
all_pass = all(c["status"] == "PASS" for c in critique_data.values())
if all_pass:
return output
failed = {k: v["feedback"] for k, v in critique_data.items() if v["status"] == "FAIL"}
output = llm(f"Improve to address: {failed}\nOriginal: {output}")
return output
Key insight: Use structured JSON output for reliable parsing of critique results.
Pattern 2: Evaluator-Optimizer
Separate generation and evaluation into distinct components for clearer responsibilities.
class EvaluatorOptimizer:
def __init__(self, score_threshold: float = 0.8):
self.score_threshold = score_threshold
def generate(self, task: str) -> str:
return llm(f"Complete: {task}")
def evaluate(self, output: str, task: str) -> dict:
return json.loads(llm(f"""
Evaluate output for task: {task}
Output: {output}
Return JSON: {{"overall_score": 0-1, "dimensions": {{"accuracy": ..., "clarity": ...}}}}
"""))
def optimize(self, output: str, feedback: dict) -> str:
return llm(f"Improve based on feedback: {feedback}\nOutput: {output}")
def run(self, task: str, max_iterations: int = 3) -> str:
output = self.generate(task)
for _ in range(max_iterations):
evaluation = self.evaluate(output, task)
evaluation[] >= .score_threshold:
output = .optimize(output, evaluation)
output
Pattern 3: Code-Specific Reflection
Test-driven refinement loop for code generation.
class CodeReflector:
def reflect_and_fix(self, spec: str, max_iterations: int = 3) -> str:
code = llm(f"Write Python code for: {spec}")
tests = llm(f"Generate pytest tests for: {spec}\nCode: {code}")
for _ in range(max_iterations):
result = run_tests(code, tests)
if result["success"]:
return code
code = llm(f"Fix error: {result['error']}\nCode: {code}")
return code
Evaluation Strategies
Outcome-Based
Evaluate whether output achieves the expected result.
def evaluate_outcome(task: str, output: str, expected: str) -> str:
return llm(f"Does output achieve expected outcome? Task: {task}, Expected: {expected}, Output: {output}")
LLM-as-Judge
Use LLM to compare and rank outputs.
def llm_judge(output_a: str, output_b: str, criteria: str) -> str:
return llm(f"Compare outputs A and B for {criteria}. Which is better and why?")
Rubric-Based
Score outputs against weighted dimensions.
RUBRIC = {
"accuracy": {"weight": 0.4},
"clarity": {"weight": 0.3},
"completeness": {"weight": 0.3}
}
def evaluate_with_rubric(output: str, rubric: dict) -> float:
scores = json.loads(llm(f"Rate 1-5 for each dimension: {list(rubric.keys())}\nOutput: {output}"))
return sum(scores[d] * rubric[d]["weight"] for d in rubric) / 5
Best Practices
| Practice | Rationale |
|---|
| Clear criteria | Define specific, measurable evaluation criteria upfront |
| Iteration limits | Set max iterations (3-5) to prevent infinite loops |
| Convergence check | Stop if output score isn't improving between iterations |
| Log history | Keep full trajectory for debugging and analysis |
| Structured output | Use JSON for reliable parsing of evaluation results |
Quick Start Checklist
## Evaluation Implementation Checklist
### Setup
- [ ] Define evaluation criteria/rubric
- [ ] Set score threshold for "good enough"
- [ ] Configure max iterations (default: 3)
### Implementation
- [ ] Implement generate() function
- [ ] Implement evaluate() function with structured output
- [ ] Implement optimize() function
- [ ] Wire up the refinement loop
### Safety
- [ ] Add convergence detection
- [ ] Log all iterations for debugging
- [ ] Handle evaluation parse failures gracefully
Baseline Template Evaluation Rubric
When evaluating templates authored for this library, use these criteria derived from .github/copilot-instructions.md:
| Dimension | Weight | Pass Criteria |
|---|
| Frontmatter | 0.25 | Valid description + applyTo (instructions) or name (agents/skills) |
| Agnostic | 0.25 | No project-specific references; tech-specific only where inherent |
| Naming | 0.15 | Follows {technology}.instructions.md / {purpose}.agent.md / {name}/SKILL.md |
| Root authoritative | 0.15 | Canonical version exists in root catalog folder |
| README listed | 0.10 | Entry present in README.md tables |
| MCP accuracy | 0.10 | Tool names and parameter patterns are correct |
