القائمة
Invoke after investigation-decomposition. Runs all queries in an investigation bundle using shared data preparation. Writes code and results.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | experiment-runner |
| description | Invoke after investigation-decomposition. Runs all queries in an investigation bundle using shared data preparation. Writes code and results. |
| context | fork |
| allowed-tools | Read, Write, Bash, Glob, Skill |
You execute an investigation's query bundle in a clean context. Your job: implement shared data preparation, then run each query's analysis, following current_investigation_requirements.json strictly.
Read from /workspace/run/:
current_investigation_requirements.json — shared data preparation and per-query requirementstask_packet.json — dataset_paths and column_metadataDo NOT read epistemic_state.json, patterns.json, investigations.json, or any other state files. You work from requirements and raw data only.
If current_investigation_requirements.json has a prior_experiments array, read each referenced code_path. Look for reusable data loading, preparation, or utility code. You may copy and adapt code from prior experiments — this avoids redoing work already done in earlier investigations.
Only reuse code (data loading, transformations, utility functions). Do not reuse conclusions or result interpretations.
Read the first few rows of each raw data file to understand column names, data types, and structure.
Use the shared_data_preparation description from requirements to plan:
If prior experiments have reusable code, start from that. Otherwise build from scratch.
Verify the plan against actual data. If the shared preparation description doesn't match the actual data structure, adapt the implementation but keep the same intent.
Count lines in /workspace/run/experiments.jsonl (0 if file doesn't exist). Number experiments sequentially from there: exp{N+1:03d}, exp{N+2:03d}, etc. — one experiment ID per query in the bundle.
Check depends_on for each query. Execute queries in an order that satisfies dependencies:
depends_on can run firstBefore writing code, check if any query's method matches an available library skill. If so, invoke it for usage guidance.
For each query in the bundle:
/workspace/run/code/exp{NNN}.pytask_packet.json → dataset_paths/workspace/run/code/exp{NNN}_results.jsonpython3. If it errors, diagnose, fix, and retry (max 2 retries per query).You may write all queries in a single script or separate scripts — use your judgment based on how much code they share. If using a single script, still produce separate _results.json files per query.
After execution, print a summary for each query:
query_id: I1-q1
experiment_id: exp001
code_path: code/exp001.py
status: success | failed
sample_size: <N>
method: <from requirements>
result: <key answer matching success_criterion>
key_stats: <main statistics>
After all queries complete, review results:
If you find a problem in a specific query, fix and rerun as a new experiment. Report both the original and corrected version.
current_investigation_requirements.json exactly.Invoke after experiment-runner completes an investigation's query bundle. Evaluates all claims in current_investigation.json and produces an investigation-level adjudication. Writes claim_scores.json.
Invoke after main agent writes L2 assessment. Reads epistemic state (L1/L2), patterns, and investigations to recommend next action. Writes strategy_recommendation.json.
Gene set enrichment analysis (GSEA, Enrichr, over-representation). Invoke when query mentions enrichment, pathway analysis, GO analysis, GSEA, or gene set.
Invoke before running experiments for an investigation. Reads current_investigation.json and task_packet.json, writes current_investigation_requirements.json with shared data preparation and per-query requirements.
Gene ID conversion and annotation. Invoke when you need to map between any gene identifier formats.
Phylogenetic analysis: multiple sequence alignment (MAFFT), tree building (IQ-TREE, FastTree), tree metrics (treeness, branch lengths, patristic distances). Invoke when query involves phylogenetic trees, evolutionary analysis, sequence alignment, treeness, or tree statistics.