Skip to content

ci: add automated skill review for SKILL.md pull requests#177

Open
popey wants to merge 1 commit into
Jeffallan:mainfrom
popey:add-skill-review-ci
Open

ci: add automated skill review for SKILL.md pull requests#177
popey wants to merge 1 commit into
Jeffallan:mainfrom
popey:add-skill-review-ci

Conversation

@popey

@popey popey commented Mar 23, 2026

Copy link
Copy Markdown
Contributor

Hullo! Thanks for merging the skill improvements earlier. This is a follow-up that adds a lightweight GitHub Action to automatically review any SKILL.md files when they're changed in a PR, using tessl skill review.

  • Triggers only on PRs that touch **/SKILL.md
  • Posts review results as a PR comment
  • Minimal permissions: pull-requests: write and contents: read

This way you and your contributors get an instant quality signal on skill changes before manual review — no signup or tokens needed.

Adds a lightweight GitHub Action that reviews any SKILL.md files changed
in a pull request using tessl skill review. Posts results as a PR comment
with minimal permissions (pull-requests: write, contents: read).
@Jeffallan

Copy link
Copy Markdown
Owner

@popey Thanks for the PR. One question before we review further: the action installs the tessl CLI via curl -fsSL https://get.tessl.io | sh (in src/install-tessl.ts), but tessl is available on npm as a pinnable package. Is there a reason the action uses curl | sh instead of npx tessl or npm install -g tessl@0.73.0?

@popey

popey commented Mar 30, 2026

Copy link
Copy Markdown
Contributor Author

Hi @Jeffallan, apologies for the slow reply! Good question — the tesslio/skill-review action has since been updated to use a dedicated tesslio/setup-tessl action for the installation step, which makes the install process more transparent and auditable. Happy to answer any other questions.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants