BB auto-optimizer
Guide someone through optimizing anything using iterative self-improvement loops, OR run autonomous optimization loops with binary eval scoring. Systematically improve prompts, skills, websites, emails, landing pages, code, ad copy, prediction strategies, or any artifact through automated eval-mutate-select cycles. Supports two modes: binary eval (batch scoring with yes/no criteria, best for soft domains like copy/prompts/UX) and scalar eval (single float metric with git revert, best for code/ML/configs). Use this skill whenever someone says: "optimize", "improve", "make better", "iterate on", "benchmark", "eval", "self-improving", "test and improve", "split test", "A/B test", "improve my prompt", "optimize performance", "increase conversion", "improve accuracy", "score higher", "this isn't good enough", "how do I make this better", "run autoresearch", "start optimization loop", "run experiments on", "improve my metric", "autoresearch loop", "autonomous optimization", "karpathy loop", or any variation of wanting to measurably improve something over time. Also trigger for setting up automated quality loops, evaluation criteria, or self-improving systems.
As a process B 70/100 · Nearly there — weak spots: inputs and preconditions, running it twice
How to improve
- Shorten the description to 1024 characters.
- 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: 11. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1172 chars, limit 1024
Process rating: all ten parameters 70/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 5 mutating operations with no state check
- 60Tools and files. Uses tools (git, node) 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. 52 steps
- 100Failures and branches. 3 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 3767 tokens
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
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)
- +3Description length 1171: 120–800 characters recommended
- -2localhost URLs: will not work for another user
- -44 reference files, but SKILL.md never points to them: the model will not open them
- +1No license
- +2Single-language instructions
- +5Description quotes 23 example trigger phrases
- +4Structure: 32 headings
- +3Step-by-step instructions: 52 items
- +3Output format is stated explicitly
- +4Has examples (27 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 58.