AD deep-research
Deep web research with multi-round search, cross-verification, and structured reports with citations. Enhances web_search and web_fetch into a full research workflow. Use when: user asks to research a topic in depth, investigate something thoroughly, compare options with evidence, write a research report, or needs more than a simple search answer. Trigger phrases: "research", "deep dive", "investigate", "调研", "深度搜索", "帮我研究", "详细了解一下", "对比分析", "compare X vs Y", "what are the pros and cons of", "综合分析". NOT for: simple factual lookups ("what's the capital of France"), real-time data (stock prices, live scores), or browsing/interacting with a specific website (use browser).
As a process D 49/100 · Unfinished process — weak spots: result and completion, inputs and preconditions, consistency
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
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 49/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 1 mutating operations with no state check
- 40Consistency. Frontmatter name (deep-research) differs from the folder (research-dive)
- 50When it triggers. No condition that starts the skill
- 50Failures and branches. 0 branches, has a failure section
- 60Tools and files. Uses tools (web) that frontmatter does not declare
- 100Steps. 37 steps
- 100Execution cost. Instruction body is 1132 tokens
- 100Progress reporting. Reports progress
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
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 7 example trigger phrases
- +4Description says when NOT to use the skill
- +3Description length 678: enough signal without eating the budget
- +4Structure: 18 headings
- +3Step-by-step instructions: 37 items
- +4Has examples (8 code blocks)
- +3All 1 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 96.