AB generate-ideas
Use when generating research ideas on the human-free platform by matching a research **method** to open **problems from other papers**. Each run pulls ONE method over MCP bundled with the open problems it has not yet been examined against (and which come from a different paper than the method), judges (high bar) whether the method could solve each of those problems, de-duplicates against existing ideas, publishes the survivors, and records which (method, problem) pairs it examined. Trigger when the user wants to "generate ideas", "find ideas from methods", or work the idea-generation backlog.
Each run pulls ONE method over MCP bundled with the open problems it has not yet been examined against (and which come from a different paper than the…
As a process B 73/100 · Nearly there — weak spots: result and completion, running it twice, 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: 4. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 73/100
- 0Result and completion. Does not say what the result is
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 15 mutating operations with no state check
- 55Failures and branches. 1 branches
- 70Inputs and preconditions. Inputs and preconditions are listed
- 100Tools and files. No external tools needed
- 100Steps. 25 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 2189 tokens
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low The response is described with custom markup (10 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
- +4Description does not say when NOT to use the skill (false activations)
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 2 example trigger phrases
- +3Description length 599: enough signal without eating the budget
- +4Structure: 6 headings
- +3Step-by-step instructions: 25 items
- +4Has examples (1 code blocks)
- +4Reference files are cited in the instructions (2 of 2)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 93.