- Home
- Skills
- Willoscar
- Research Units Pipeline Skills
- Workspace Init
workspace-init_skill
- Python
109
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill willoscar/research-units-pipeline-skills --skill workspace-init- SKILL.md2.5 KB
Overview
This skill initializes a new artifact-first workspace by copying a standard template into a target directory. It creates the core files and folders used by research pipelines without modifying the bundled template assets. The default behavior avoids overwriting existing files unless an explicit overwrite flag is provided.
How this skill works
Given a target workspace path, the skill creates the directory if missing and copies the template contents from the bundled assets folder into the workspace. It ensures core artifacts exist (STATUS.md, CHECKPOINTS.md, UNITS.csv, DECISIONS.md) and populates standard folders (papers, outline, citations, output) with placeholder files. Existing files are preserved by default; --overwrite enables replacement.
When to use it
- Before running any pipeline that requires a workspace skeleton and artifact files
- When starting a new research run or experiment that follows the evidence-first workflow
- When you need a consistent project layout for units, checkpoints, decisions, and status tracking
- When onboarding contributors so everyone starts from the same template structure
Best practices
- Initialize workspaces under a dedicated workspaces/ directory rather than using the repository root
- Do not re-run with --overwrite unless you intend to replace template files in the workspace
- Verify UNITS.csv after initialization to ensure it has the required header and valid CSV format
- Treat the bundled assets folder as read-only; do not edit the template assets in place
- Use the created STATUS.md and CHECKPOINTS.md to record progress and acceptance evidence from the start
Example use cases
- Create a fresh workspace for a new research experiment: python .codex/skills/workspace-init/scripts/run.py --workspace workspaces/my-run
- Reinitialize a workspace and intentionally replace template files: python .codex/skills/workspace-init/scripts/run.py --workspace workspaces/my-run --overwrite
- Prepare a shared project skeleton for collaborators so everyone has STATUS.md, UNITS.csv, and the standard folders
- Recover missing artifact files by running the init on an existing workspace without --overwrite to add only missing files
FAQ
Existing files are preserved by default. Re-run with --overwrite only if you want to replace template files.
Can this modify the template assets?
No. The skill never modifies the bundled assets folder; treat that folder as read-only.
What core files must be present after initialization?
STATUS.md, CHECKPOINTS.md, UNITS.csv, and DECISIONS.md must exist. The script also creates GOAL.md, queries.md and standard folders with placeholders.