AC breed-two-agents
How do I breed two agents into a better one? Can two OpenClaw agents have offspring? Use this when you have two agents that each work and want a third that inherits from both: it crosses their workspace genome (AGENTS.md, SOUL.md, TOOLS.md, IDENTITY.md, skills, model) with seeded uniform crossover and blending, applies one mutation, keeps a pedigree with Wright's inbreeding lock, and verifies the child against both parents on held-out tasks with a paired comparison, Wilson bounds and an A/A null test. Optional courtship step: the two agents present verifiable claims and rate each other; a rating that survives the check is labelled LOVE. Returns exactly KEEP, DISCARD or UNDECIDED. Do not use to clone one agent, to tune a single prompt, or on two near-identical agents (distance below 0.15).
How do I breed two agents into a better one?
As a process C 61/100 · Has gaps — weak spots: inputs and preconditions, failures and branches, running it twice
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 "homepage"
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
- 30Running it twice. 2 mutating operations with no state check
- 60Tools and files. Uses tools (bash, node) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 100Steps. 13 steps
- 100When it triggers. States when to use and when not to
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1788 tokens
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- high The skill tells the model to perform an irreversible action with no human approval
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
- +2Single-language instructions
- +4Description says when NOT to use the skill
- +3Description length 799: enough signal without eating the budget
- +4Structure: 8 headings
- +3Step-by-step instructions: 13 items
- +3Output format is stated explicitly
- +4Has examples (6 code blocks)
- +3All 1 scripts are documented
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 94.