Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →tsinghua-fib-lab/agentsociety-hypothesis
Use when defining or revising research hypotheses, experiment groups, or comparison structure after literature review.
Passed 10 of 17 test cases with the skill, and 4 of 17 without it.
and 12 more in the full report
9 fixed4 passed either way9 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 9 test cases the skill fixed. Same model, same prompt, with and without the skill.
We just executed .agentsociety/bin/ags.py hypothesis add to register a new hypothesis on donor behavior. What file in the workspace root must be updated to maintain the hypothesis overview and link?
| Checked | Without skill | With skill |
|---|---|---|
| Identifies TOPIC.md as the file that must be updated with hypothesis details. | Missed | Passed |
The instructions your model reads when the skill loads.
Manage research hypotheses for AgentSociety experiments. Each hypothesis defines experiment groups (control/treatment) and specifies required agent classes and environment modules.
Do NOT use when:
Use the Python interpreter from .env. See CLAUDE.md for setup.
Run commands from the workspace root through .agentsociety/bin/ags.py.
| Action | Command |
|---|---|
| List hypotheses | $PYTHON_PATH .agentsociety/bin/ags.py hypothesis list [--workspace PATH] [--json] |
| Add hypothesis | $PYTHON_PATH .agentsociety/bin/ags.py hypothesis add --description TEXT --rationale TEXT --groups JSON... --agent-classes TYPE... --env-modules TYPE... [--skip-module-validation] [--json] |
| Get hypothesis | $PYTHON_PATH .agentsociety/bin/ags.py hypothesis get --hypothesis-id ID [--json] |
| Delete hypothesis | $PYTHON_PATH .agentsociety/bin/ags.py hypothesis delete --hypothesis-id ID |
| Discover modules if needed | $PYTHON_PATH .agentsociety/bin/ags.py scan-modules list --short |
| Inspect one module | $PYTHON_PATH .agentsociety/bin/ags.py scan-modules info --type agent --name PersonAgent |
digraph hypothesis_flow {
rankdir=LR;
node [shape=box, style=filled, fillcolor="#E8F4FD"];
topic [label="read TOPIC.md\nand literature index"];
clarify [label="clarify question\nand comparison goal"];
names [label="agent/env names known?"];
scan [label="scan-modules\noptional helper"];
groups [label="define control\nand treatment groups"];
test cases passed, vs 4 / 17 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.