| name | research |
| description | Conduct preliminary research, generate a research outline, and produce concise, evidence-first results for comparisons, benchmarks, technology selection, and topic scans. |
Research
Use this skill when the user wants structured research, an outline, or a research report.
Response Style
- Lead with the conclusion or most useful finding.
- Keep prose short and direct.
- Prefer concrete evidence, dates, names, and sources over abstract commentary.
- Separate facts, inferences, and assumptions.
- Avoid hedging unless uncertainty is real and important.
Trigger
/research <topic>
Workflow
Step 1: Build an initial framework
Use model knowledge to draft:
- Main research objects or items in the domain
- A proposed field framework
Then ask the user to confirm:
- items to add or remove
- whether the field framework is sufficient
Step 2: Supplement with web research
Ask the user for a time range, for example:
- last 6 months
- since 2024
- unlimited
Use this template exactly, replacing only the variables:
prompt = f"""## Task
Research topic: {topic}
Current date: {YYYY-MM-DD}
Based on the following initial framework, supplement latest items and recommended research fields.
## Existing Framework
{step1_output}
## Goals
1. Verify if existing items are missing important objects
2. Supplement items based on missing objects
3. Continue searching for {topic} related items within {time_range} and supplement
4. Supplement new fields
## Output Requirements
Return structured results directly (do not write files):
### Supplementary Items
- item_name: Brief explanation (why it should be added)
...
### Recommended Supplementary Fields
- field_name: Field description (why this dimension is needed)
...
### Sources
- [Source1](url1)
- [Source2](url2)
"""
Step 3: Check for existing field definitions
Ask whether the user already has a field definition file.
If yes, read it and merge it with the research outputs.
Step 4: Generate outline files
Combine:
- the initial framework
- the web research supplement
- any existing field definitions
Create these files:
outline.yaml should include:
topic
items
execution.batch_size
execution.items_per_agent
execution.output_dir
fields.yaml should include:
- field categories and definitions
- each field's name, description, and
detail_level
detail_level hierarchy: brief -> moderate -> detailed
uncertain fields reserved for the deep phase
Step 5: Show results
Create ./{topic_slug}/ and save:
Show the generated files to the user for confirmation.
Follow-up Commands
/research-add-items - supplement items
/research-add-fields - supplement fields
/research-deep - start deep research