Upskill

UPskill

Generate and evaluate agent skills based on traces with agents. Create skills with teacher models (expensive/slow) that student models (cheap/fast) can use to perform harder tasks reliably.

[!TIP]

UPskill v2 - recommended default config file now runs evaluations on Hugging Face Jobs. Make sure to set your `HF_TOKEN` and use `--artifact-repo ` for job creation and result capture

Quick Start

Install upskill:

uv pip install upskill
# or just use uv
uvx upskill

Create a new skill

upskill generate "write good git commit messages"
# or based on previous agent traces
upskill generate "document the pattern" --from ./trace.md
# Skills are saved to ./skills/{skill-name}/ by default

Generate a skill with a teaching model and evaluate it on a student model.

upskill generate "write good git commit messages" --model sonnet --eval-model haiku

Benchmark a set of models against a skill.

upskill eval ./skills/git-commit-messages/ -m haiku -m sonnet
# logs pretty printed to the terminal

View the results later.

upskill runs --skill git-commit-messages

Development checks

This repo uses a CI flow inspired by `fast-agent` with separate format, lint, typecheck, and test stages.

Install dev dependencies:

uv sync --extra dev

Run the quality gates locally:

uv run scripts/format.py
uv run scripts/lint.py
uv run scripts/typecheck.py
uv run scripts/cpd.py --check
uv run --extra dev pytest -v

Or use the helper script to run the whole sequence:

uv run scripts/check.py

Add `--sync` to include `uv sync --extra dev`, or `--skip-tests` for a faster static-only pass.

To auto-format before re-running checks:

uv run --extra dev scripts/format.py --write

Current enforced standards:

• `ruff f