AB autoimprove
Autonomous optimization loop that improves any measurable thing. Point it at files to change, a command to check, and a number to improve — then walk away. Works with any AI agent. Use when the user wants to autonomously optimize code performance, ML training, Docker images, SQL queries, prompts, CI speed, bundle size, Kubernetes configs, or any domain with a measurable score. Triggers include requests like 'optimize this', 'improve performance', 'make this faster', 'reduce allocations', 'autoimprove', 'run the optimization loop', 'let it run overnight', or when the user has an improve.md file.
As a process B 77/100 · Nearly there — weak spots: execution cost, running it twice
How to improve
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- 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: 6. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 9034 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 77/100
- 30Running it twice. 17 mutating operations with no state check
- 40Execution cost. Instruction body is 9034 tokens: crowds the task out of the window
- 60Tools and files. Uses tools (bash, git, python) 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. 121 steps
- 100Inputs and preconditions. Inputs and preconditions are listed
- 100Failures and branches. 5 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 13 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (7 tags): a typed call is more reliable
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
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- -2localhost URLs: will not work for another user
- +1No license
- +2Single-language instructions
- +3Description length 601: enough signal without eating the budget
- +4Structure: 37 headings
- +3Step-by-step instructions: 121 items
- +3Output format is stated explicitly
- +4Has examples (21 code blocks)
- +4Reference files are cited in the instructions (2 of 2)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 79.