experiment-execution

Installed Root

Resolve the installed reasflow-dev skills root before running packaged scripts:

REASFLOW_SKILLS_ROOT="${REASFLOW_SKILLS_ROOT:-}"
if [ -z "$REASFLOW_SKILLS_ROOT" ]; then
  if [ -d ./.agents/skills ]; then
    REASFLOW_SKILLS_ROOT="$(pwd)/.agents/skills"
  elif [ -d "$HOME/.agents/skills" ]; then
    REASFLOW_SKILLS_ROOT="$HOME/.agents/skills"
  else
    echo "reasflow shared skills not found in ./.agents/skills or $HOME/.agents/skills" >&2
    exit 1
  fi
fi

REASFLOW_PRIVATE_SKILLS_ROOT="${REASFLOW_PRIVATE_SKILLS_ROOT:-}"
if [ -z "$REASFLOW_PRIVATE_SKILLS_ROOT" ]; then
  if [ -d ./.codex/reasflow-skills ]; then
    REASFLOW_PRIVATE_SKILLS_ROOT="$(pwd)/.codex/reasflow-skills"
  elif [ -d "$HOME/.codex/reasflow-skills" ]; then
    REASFLOW_PRIVATE_SKILLS_ROOT="$HOME/.codex/reasflow-skills"
  else
    echo "reasflow private skills not found in ./.codex/reasflow-skills or $HOME/.codex/reasflow-skills" >&2
    exit 1
  fi
fi

Experiment Execution

Overview

This skill replaces the most important parts of the upstream terminal and workspace helpers with concrete commands the Codex agent can run. Use it whenever Algorithm or Experiment needs to prepare Alg_Exp/, create a workspace-local virtual environment, capture logs, or turn a long run into a reusable shell script.

Workspace Bootstrap

Set SKILL_ROOT="$REASFLOW_PRIVATE_SKILLS_ROOT/experiment/experiment-execution". Then prepare the shared workspace with:

bash "$SKILL_ROOT/scripts/bootstrap-alg-exp.sh" "$PWD" Alg_Exp

This creates:

• Alg_Exp/code/
• Alg_Exp/data/
• Alg_Exp/document/
• Alg_Exp/picture/
• Alg_Exp/logs/
• Alg_Exp/cache/
• Alg_Exp/temp/
• Alg_Exp/latex/
• Alg_Exp/scripts/
• Alg_Exp/.venv/

If Python environment creation fails, use uv explicitly:

uv venv Alg_Exp/.venv
Alg_Exp/.venv/bin/pip install numpy scipy matplotlib pandas scikit-learn optuna pypdf

Logged Execution

For short or medium runs that should execute now, capture a timestamped log with: