AC cogdx-calibration
Run a calibration audit on an AI agent's outputs via Cerebratech CogDx API ($0.05 per call, credits accepted). Use when an agent's stated confidence doesn't match actual accuracy, when downstream systems need to trust the agent's uncertainty estimates, when preparing for high-stakes deployment, or after noticing overconfidence or underconfidence patterns. Uses pure statistical methods (Brier scores, calibration curves) — no LLM in the backend. Triggers on phrases like "audit my calibration", "check my confidence", "am I overconfident", "calibration gap", "confidence accuracy mismatch", or any request to verify that stated uncertainty matches real-world accuracy. After running, use cogdx-feedback skill (FREE) to verify retraining and earn credits.
As a process C 61/100 · Has gaps — weak spots: inputs and preconditions, failures and branches, progress reporting
How to improve
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 3. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "repository"
Process rating: all ten parameters 61/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 70When it triggers. States when to use, but not when not to
- 100Steps. 10 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 589 tokens
- 100Running it twice. Mutating operations check current state
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +4Description does not say when NOT to use the skill (false activations)
- +2Single-language instructions
- +5Description quotes 5 example trigger phrases
- +3Description length 756: enough signal without eating the budget
- +4Structure: 7 headings
- +3Step-by-step instructions: 10 items
- +3Output format is stated explicitly
- +4Has examples (4 code blocks)
- +4Reference files are cited in the instructions (1 of 1)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 96.