Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →tsinghua-fib-lab/agentsociety-run-experiment
Use when experiment configuration already exists and a simulation run needs to be started, monitored, or stopped.
Passed 15 of 19 test cases with the skill, and 3 of 19 without it.
and 15 more in the full report
13 fixed4 passed either way5 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 13 test cases the skill fixed. Same model, same prompt, with and without the skill.
I am running a multi-project setup where the project workspace root is at `/data/project_alpha` instead of the current working directory. Which CLI argument sets this workspace root path?
| Checked | Without skill | With skill |
|---|---|---|
| Includes the --workspace flag with /data/project_alpha. | Missed | Passed |
The instructions your model reads when the skill loads.
Manage experiment execution using the AgentSociety2 CLI: start, monitor, and stop simulation runs.
init_config.json and steps.yaml exist and the user wants to start a simulationDo NOT use when:
Run commands from the workspace root through .agentsociety/bin/ags.py.
| Action | Command |
|---|---|
| Start | $PYTHON_PATH .agentsociety/bin/ags.py run-experiment start --hypothesis-id ID --experiment-id ID |
| Status | $PYTHON_PATH .agentsociety/bin/ags.py run-experiment status --hypothesis-id ID --experiment-id ID |
| Stop | $PYTHON_PATH .agentsociety/bin/ags.py run-experiment stop --hypothesis-id ID --experiment-id ID |
| List | $PYTHON_PATH .agentsociety/bin/ags.py run-experiment list |
Use the Python interpreter from .env. See CLAUDE.md for setup.
digraph run_experiment_flow {
rankdir=LR;
node [shape=box, style=filled, fillcolor="#E8F4FD"];
check [label="confirm init_config.json\nand steps.yaml"];
start [label="start"];
status [label="status / logs"];
done [label="run/replay/_schema.json\nand artifacts present"];
stop [label="stop"];
fix [label="inspect logs\nand revise config"];
test cases passed, vs 3 / 19 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.