sim_brief — Claude Code prompt: simulation and setup
The sim_brief Claude Code prompt guides a clinical-trial simulation and its report. It runs when an agent is asked to implement a model from params.json.
no license ★ 654 updated 2026-11-09
/plugin marketplace add anthropics/life-sciences
/plugin install clinical-trial-forecast@life-sciencesWhat sim_brief does
The prompt directs an agent to build a clinical-trial simulation and prepare its outputs. Values in params.json govern the implementation; any ambiguous interpretation must be recorded in the run log and the report’s implementation-choices table.
The requested files include simulate.py, results.json, PNG figures, a report chapter in sim_report.md, and run_log.txt with a two-seed reproducibility check. The simulation script uses numpy, scipy, and Python’s standard library; figures require matplotlib. Running it requires python3 and bg.py. The specified run must finish in under 60 minutes.
What it helps with
- implement a simulation from
params.json - produce results and figures for a report
- record decisions on ambiguous parameters
- check reproducibility with two seeds
Example prompts for the agent
- Build the clinical-trial simulation from
params.jsonand prepare all report files. - Implement the specified model and log decisions about ambiguous parameters.
- Run the requested number of simulations and check reproducibility with two seeds.
How to install sim_brief
1Plugin marketplace
/plugin marketplace add anthropics/life-sciences/plugin install clinical-trial-forecast@life-sciences2Manually: put the files in place
.claude/commands/sim_brief.mdHow to use
Paste the text at the start of a chat or into project instructions
Questions about sim_brief
How do I install sim_brief in Claude Code?
Run /plugin marketplace add anthropics/life-sciences, then /plugin install clinical-trial-forecast@life-sciences. The prompt file is at .claude/commands/sim_brief.md.
What does sim_brief do?
It guides implementation of a clinical-trial simulation and preparation of the code, results, figures, report, and run log.
What does sim_brief require?
It requires params.json, python3, bg.py, and the numpy, scipy, and matplotlib libraries. The simulation script uses numpy, scipy, and Python’s standard library; matplotlib is used for figures.