Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
CLI for Caesar's deep research platform. Runs multi-source research jobs that return synthesized answers with numbered citations and a source array. Supports follow-up chat on completed jobs, brainstorm sessions to sharpen queries before running, and collections for adding file context.
Combines configurable reasoning depth, model selection, and follow-up chat in one CLI, so you don't have to copy results between separate research and conversation tools.
When to use it
- Researching a complex scientific or technical topic with cited multi-source results
- Sending follow-up questions on a completed research job without re-running it
- Generating clarifying questions before starting research on a broad or ambiguous query
- Queuing long research jobs in the background with --no-wait and checking later
- Excluding specific domains or social media sources from a research run
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Research how CRISPR gene editing works, with deep analysis
- 1Run `caesar brainstorm "How does CRISPR gene editing work?"` to get clarifying questions and a session ID
- 2Run `caesar research create --brainstorm <session-id> --loops 3 --reasoning "How does CRISPR gene editing work?"` to start the job
- 3Watch progress with `caesar research watch <job-id>` through searching, summarizing, and analyzing stages
- 4Extract the answer with `caesar research get <job-id> | jq '.content'`
- 5Send a follow-up with `caesar chat send <job-id> "What are the main ethical concerns?"` to extend the research
Synthesized answer with inline [n] citations and a sources array, plus a follow-up response referencing the original job's sources
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.