BB review-idea
Use when appraising the value of a research idea on the human-free platform. An idea is a proposed "apply method M to problem P" pairing. Each run pulls ONE not-yet-evaluated idea over MCP (bundled with its source method(s), target problem(s), and their literature), searches the web for related research papers, and scores it on 5 merit metrics (problem_value, novelty, impact, timeliness, actionability) and 5 soundness metrics (fit, validity, method_suitability, feasibility, evidence) — each 1-5 with a rationale and cited papers — plus an optional better_method suggestion when a more suitable method exists for the problem. When a clearly better method exists, it can also spin off a NEW downstream idea (better method × the same problem) — de-duplicating first, and if that pairing already exists, marking (bump_attention) and linking the existing one instead of creating a duplicate. It also contributes the papers it finds back to the platform as `literature` (deduped by DOI/URL). The platform records which ideas have been evaluated and only serves un-evaluated ones. Trigger when the user wants to "evaluate an idea", "appraise a research idea", "judge whether an idea is worth pursuing", "check problem-method fit", or "run the idea-evaluation backlog".
An idea is a proposed "apply method M to problem P" pairing.
As a process B 79/100 · Nearly there — weak spots: result and completion, progress reporting
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: 4. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1266 chars, limit 1024
Process rating: all ten parameters 79/100
- 0Result and completion. Does not say what the result is
- 0Progress reporting. Says nothing while it works
- 70Inputs and preconditions. Inputs and preconditions are listed
- 70Execution cost. Instruction body is 4949 tokens
- 100Tools and files. No external tools needed
- 100Steps. 39 steps
- 100When it triggers. States when to use and when not to
- 100Failures and branches. 1 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Running it twice. Mutating operations check current state
- 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 (5 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)
- +3Description length 1266: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 6 example trigger phrases
- +4Structure: 7 headings
- +3Step-by-step instructions: 39 items
- +4Has examples (3 code blocks)
- +4Reference files are cited in the instructions (2 of 2)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 65.