diff --git a/.github/workflows/pages.yml b/.github/workflows/pages.yml new file mode 100644 index 0000000..d9c138e --- /dev/null +++ b/.github/workflows/pages.yml @@ -0,0 +1,37 @@ +# Deploy the official website (website/) to GitHub Pages +name: Deploy Pages + +on: + push: + branches: [main] + paths: ['website/**', '.github/workflows/pages.yml'] + workflow_dispatch: + +permissions: + contents: read + pages: write + id-token: write + +concurrency: + group: pages + cancel-in-progress: true + +jobs: + build: + runs-on: ubuntu-latest + steps: + - uses: actions/checkout@v4 + - name: Upload Pages artifact + uses: actions/upload-pages-artifact@v3 + with: + path: website + + deploy: + needs: build + runs-on: ubuntu-latest + environment: + name: github-pages + url: ${{ steps.deployment.outputs.page_url }} + steps: + - id: deployment + uses: actions/deploy-pages@v4 diff --git a/website/index.html b/website/index.html new file mode 100644 index 0000000..542e5e7 --- /dev/null +++ b/website/index.html @@ -0,0 +1,293 @@ + + +
+ + ++ Heimdall is an AI code review bot for GitHub. Model-agnostic, self-hostable, and always on call β + it reviews every PR with severity grading, inline comments, and executable diff suggestions. +
+ +Model freedom, private deployment, and reports you can actually act on.
+Claude, GPT, Gemini, or a local Ollama/vLLM β you pick. One unified AI_API_KEY + AI_BASE_URL.
Every issue pinned to its code line, with an actionable title, a fix suggestion, and an executable diff (even 1-Click Suggestions).
PRs stay quiet until you say the word β comment @CoderHeimdall. Or set auto_review: true.
Trust-boundary, sensitive-field leakage, fake implementations, N+1, swallowed exceptions β caught and graded.
REVIEW_LANGUAGE = en (default), zh, or bilingual. One bot, every team.
.github/heimdall.yml β include/exclude files, min severity, custom instructions, whitelist, block-on-critical.
GitHub Actions (per-repo), Cloudflare Workers (installable App), or Probot self-hosted β code stays on your infra.
Triple dedup means one review per commit. No duplicate review dumps.
Change summary, severity table, focus areas, verification steps β then pinned inline comments with diffs.
+One fine morning, Copilot arbitrarily revised its subscription deal: models quietly downgraded, token limits shrank, choices vanished overnight. And that $10/month fee? Collected right on schedule β yet your Code Review quota reliably ran out by mid-month, greeting you with "Quota limit reached. Please upgrade."
+By month's end: paid full price, ran out of tokens halfway through, zero Code Review when PRs hit β and helped another earnings report.
+Pick a mode β Actions for a single repo, Workers for a team-wide installable bot.
+# Mode A β add an AI reviewer to any repo +$ mkdir -p .github/workflows scripts +$ cp template/heimdall-review.yml .github/workflows/ +$ cp scripts/heimdall-review.js scripts/ +$ # add AI_API_KEY (or ANTHROPIC_API_KEY) as a repo secret + +# in any PR, comment: +@CoderHeimdall + +# want auto-review? add .github/heimdall.yml with: +auto_review: true+ +