Enhanced GitHub Action to run OpenCode GitHub agent

Create .github/workflows/opencode.yml in the repository where you want OpenCode to respond to issue and pull request comments. By default, the action exchanges the workflow OIDC token for an OpenCode GitHub App token, so the workflow must grant id-token: write.

---
name: OpenCode
on:
  issue_comment:
    types:
      - created
  pull_request_review_comment:
    types:
      - created
permissions:
  contents: write
  issues: write
  pull-requests: write
  id-token: write
jobs:
  opencode:
    if: contains(github.event.comment.body, '/oc') || contains(github.event.comment.body, '/opencode')
    runs-on: ubuntu-latest
    steps:
      - name: Checkout repository
        uses: actions/checkout@v7
        with:
          fetch-depth: 1
          persist-credentials: false
      - name: Run OpenCode
        uses: dceoy/opencode-action@main
        env:
          OPENCODE_API_KEY: ${{ secrets.OPENCODE_API_KEY }}
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
        with:
          model: opencode-go/glm-5.2

Then comment /opencode or /oc on an issue, pull request, or pull request review comment.

Set the API key required by the selected model provider, for example:

• OPENCODE_API_KEY for OpenCode models
• OPENROUTER_API_KEY for OpenRouter models
• ANTHROPIC_API_KEY for Anthropic models
• OPENAI_API_KEY for OpenAI models

When use-github-token: true, pass GITHUB_TOKEN in env and grant the workflow enough permissions for the requested work.

The bundled /review-pr command produces a single markdown review response that the OpenCode GitHub integration posts to the PR as opencode-agent[bot]. The command does not post to GitHub directly, so a review run yields one OpenCode PR comment rather than a separate github-actions[bot] review. Workflows that invoke it must provide:

• GH_TOKEN: ${{ github.token }} or GITHUB_TOKEN: ${{ github.token }} for gh pr diff / gh pr view reads
• A valid API key for the selected model provider with available credits or quota

Example OpenCode step:

- name: Run OpenCode review
  uses: dceoy/opencode-action@main
  env:
    OPENROUTER_API_KEY: ${{ secrets.OPENROUTER_API_KEY }}
    GH_TOKEN: ${{ github.token }}
  with:
    model: openrouter/openrouter/free
    prompt: /review-pr

The bundled toolkit combines Claude Code Action-style core reviewers with pr-review-toolkit specialty agents. Use /review-pr with any of the following aspect keywords:

• code-quality-reviewer — general quality, maintainability, edge cases, robustness, type safety
• test-coverage-reviewer — missing critical test scenarios, brittle tests, error coverage gaps
• documentation-accuracy-reviewer — README, API docs, docstrings, examples vs. implementation

• code-reviewer — project-guideline compliance (AGENTS.md), bugs, and quality
• performance-reviewer — algorithmic complexity, N+1, resource leaks
• security-code-reviewer — trust boundaries, injection, secrets, auth/authz
• pr-test-analyzer — behavioral test coverage and critical coverage gaps
• silent-failure-hunter — silent failures, broad catch blocks, fallback logic
• comment-analyzer — comment accuracy, completeness, and comment rot
• type-design-analyzer — type invariants, encapsulation, and design quality

Findings are normalized, deduplicated across agents, and validated against the PR diff before being returned. All findings go into a single markdown response as file:line references; duplicated root causes produce one summary entry listing all affected files, and cross-document or cross-file consistency problems are summarized once instead of repeated at each location. The surrounding opencode github run integration posts that response to the PR as opencode-agent[bot].

Explain an issue:

/opencode explain this issue

Fix an issue and open a pull request:

/opencode fix this

Request a change on a pull request:

Delete the attachment from S3 when the note is removed /oc

Request a change on specific code lines from the pull request Files tab:

/oc add error handling here