# Gadfly — agentic adversarial PR reviewer (https://gitea.stevedudenhoeffer.com/steve/gadfly). # # Runs the published Gadfly image (pinned to an immutable :sha- tag — act_runner # caches :latest, and this build is what carries foreman provider-type support) # as a specialist swarm and posts # ONE consolidated review comment as gitea-actions. Advisory only — never blocks a # merge. Gadfly reviewing its OWN PRs — dogfooding: 6 cloud models + the Claude # Code engine (sonnet) as a competitor. Local Macs and weak cloud models dropped. name: Adversarial Review (Gadfly) on: pull_request: types: [opened, reopened, ready_for_review] issue_comment: types: [created] workflow_dispatch: inputs: pr_number: description: "PR number to review" required: true permissions: contents: read issues: write pull-requests: write concurrency: group: gadfly-${{ github.event.issue.number || github.event.pull_request.number || github.event.inputs.pr_number }} cancel-in-progress: true jobs: review: # Security: only trusted users may trigger a secret-bearing run via a PR # comment (pull_request + workflow_dispatch are already trusted). Mirrors # GADFLY_ALLOWED_USERS, the in-container belt-and-suspenders check. if: >- github.event_name != 'issue_comment' || (github.event.issue.pull_request && (github.actor == 'steve' || github.actor == 'fizi' || github.actor == 'dazed')) runs-on: ubuntu-latest # Fleet (6 cloud + claude-code/sonnet, all concurrent) reviewing every PR # with the 3-lens suite. All cloud now, so runs are fast. timeout-minutes: 90 steps: - uses: docker://gitea.stevedudenhoeffer.com/steve/gadfly:sha-80d8f53 env: GITEA_API: ${{ github.server_url }}/api/v1/repos/${{ github.repository }} GITEA_TOKEN: ${{ secrets.GITEA_TOKEN }} OLLAMA_CLOUD_API_KEY: ${{ secrets.OLLAMA_CLOUD_API_KEY }} # Claude Code engine auth (for the claude-code/* entries in GADFLY_MODELS # below): Pro/Max subscription token. Dogfoods the Phase-1 engine on # gadfly's own PRs as a competitor alongside the Ollama models. CLAUDE_CODE_OAUTH_TOKEN: ${{ secrets.CLAUDE_CODE_OAUTH_TOKEN }} # Fleet: 6 cloud (3 at a time) + Claude Code (sonnet + opus) — one # consolidated comment each, all cloud now. The local Macs (m1/m5) and # the weaker cloud models (gemma4, gpt-oss:120b, kimi-k2.7-code) were # dropped as low-signal. The claude-code/* entries run the Phase-1 # engine as competitors in their own lane (need CLAUDE_CODE_OAUTH_TOKEN). GADFLY_MODELS: "minimax-m3:cloud,glm-5.2:cloud,glm-5.1:cloud,deepseek-v4-pro:cloud,nemotron-3-super:cloud,qwen3-coder:480b-cloud,claude-code/sonnet,claude-code/opus" # cloud runs 3 at once; claude-code 2 at a time; both lanes parallel. GADFLY_PROVIDER_CONCURRENCY: "ollama-cloud=3,claude-code=2" # 3 cloud models x 3 lenses = 9 concurrent ollama-cloud queries (under the 10 budget). GADFLY_PROVIDER_LENS_CONCURRENCY: "ollama-cloud=3" # Default => the 3-lens suite (security, correctness, error-handling). # Set the repo var GADFLY_SPECIALISTS to override (csv / "all" / "auto"). GADFLY_SPECIALISTS: ${{ vars.GADFLY_SPECIALISTS || 'security,correctness,error-handling' }} # Per-lens deadline + bounded steps (also bounds the claude-code lane). GADFLY_TIMEOUT_SECS: "600" GADFLY_MAX_STEPS: "14" # Allow-list for the comment trigger (mirrors the job-level if: guard). GADFLY_ALLOWED_USERS: "steve,fizi,dazed" # --- findings telemetry: POST runs + findings to the gadfly-reports store --- # Advisory & off unless GADFLY_FINDINGS_URL is set; failures only log to # stderr and never affect the review. GADFLY_REPO / GADFLY_PR are derived # in-container; the URL + token are user-scope secrets. GADFLY_FINDINGS_URL: ${{ secrets.GADFLY_FINDINGS_URL }} GADFLY_FINDINGS_TOKEN: ${{ secrets.GADFLY_FINDINGS_TOKEN }} # --- event context (leave as-is) --- EVENT_NAME: ${{ github.event_name }} PR: ${{ github.event.pull_request.number || github.event.issue.number || github.event.inputs.pr_number }} PR_BRANCH: ${{ github.head_ref }} IS_DRAFT: ${{ github.event.pull_request.draft }} COMMENT_BODY: ${{ github.event.comment.body }} COMMENT_ID: ${{ github.event.comment.id }} ACTOR: ${{ github.actor }}