Skillgrade Grader Authoring

Procedures

Step 1: Identify the Grading Strategy

1. Determine whether the task requires objective verification (deterministic) or qualitative assessment (LLM rubric).
2. For most tasks, combine both: deterministic graders verify outcomes (weight 0.7), LLM rubrics assess approach quality (weight 0.3).

Step 2: Write a Deterministic Grader

1. Create a script in the skill's graders/ directory (bash or TypeScript).
2. The script must output a JSON object to stdout with the following structure:
   jsonCopy

   {"score": 0.67, "details": "2/3 checks passed", "checks": [{"name": "check-name", "passed": true, "message": "Description"}]}
   

3. score (0.0–1.0) and details are required. checks is optional but recommended.
4. Read references/grader-output-schema.md for the full output specification.
5. Use awk for arithmetic in bash scripts — bc is not available in node:20-slim.
6. Reference the grader in eval.yaml:
   yamlCopy

   - type: deterministic
     run: bash graders/check.sh
     weight: 0.7
   

Step 3: Write an LLM Rubric Grader

1. Draft a rubric with explicit scoring criteria and point allocations.
2. Structure the rubric into weighted sections that sum to 1.0:
   Copy

   Workflow Compliance (0-0.5):
   - Did the agent follow the mandatory workflow steps?
   Efficiency (0-0.5):
   - Completed in ≤5 commands without trial-and-error?
   

3. Reference the rubric in eval.yaml:
   yamlCopy

   - type: llm_rubric
     rubric: |
       [rubric text or file path]
     weight: 0.3
     model: gemini-2.0-flash  # optional, auto-detected from API key
   

4. For long rubrics, store in a separate file and reference by path: rubric: rubrics/quality.md.

Step 4: Combine Multiple Graders

1. Assign weights to each grader based on importance. Weights are normalized automatically.
2. Final reward is calculated as: Σ (grader_score × weight) / Σ weight.
3. Example configuration:
   yamlCopy

   graders:
     - type: deterministic
       run: bash graders/check.sh
       weight: 0.7
     - type: llm_rubric
       rubric: rubrics/quality.md
       weight: 0.3
   

Step 5: Validate Graders

1. Create a reference solution script that produces the expected output.
2. Run skillgrade --validate to verify graders score the reference solution correctly.
3. Test only deterministic graders: skillgrade --grader=deterministic (skips LLM calls, faster iteration).
4. Test only LLM rubric graders: skillgrade --grader=llm_rubric.
5. Run a specific eval with a specific grader type: skillgrade --eval=my-eval --grader=deterministic.
6. If a grader returns unexpected scores, inspect the script output and adjust scoring logic.

Error Handling

• If a deterministic grader outputs non-JSON, ensure all echo/console.log statements except the final JSON result are redirected to stderr.
• If an LLM rubric grader returns 0.00 with "No API key," set GEMINI_API_KEY or ANTHROPIC_API_KEY in the environment.
• If scores are inconsistent across trials, reduce rubric ambiguity by adding concrete examples of passing and failing behavior.