feat: Phase 4 — reusable "subscribe" workflow (+ dogfood it) (#8)
Build & push image / build-and-push (push) Successful in 9s

Centralizes the consumer stub into a reusable Gitea workflow
(.gitea/workflows/review-reusable.yml, workflow_call + defaulted inputs +
secrets: inherit); gadfly's own dogfood is now a thin caller of it, which
proved end-to-end that github.event context propagates into the reusable
on this act_runner. Adds the slim examples/reusable.yml stub + docs.

Folded in the swarm's findings: timeout_minutes default 30->45, map
GADFLY_API_KEY, explicit permissions block, drop the dead specialist_suite
input, and harden the example's actor gate. ~70 findings graded.

Completes the gadfly-games build (Phases 1-4 + quality fixes).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Steve Dudenhoeffer <steve@stevedudenhoeffer.com>
Co-committed-by: Steve Dudenhoeffer <steve@stevedudenhoeffer.com>
This commit was merged in pull request #8.
This commit is contained in:
2026-06-27 23:42:01 +00:00
committed by steve
parent b23eeb8cbf
commit 5f86062a5a
6 changed files with 172 additions and 59 deletions
+18 -55
View File
@@ -1,11 +1,10 @@
# Gadfly — agentic adversarial PR reviewer (https://gitea.stevedudenhoeffer.com/steve/gadfly). # Gadfly reviewing its OWN PRs — now a thin CALLER of the reusable workflow
# (.gitea/workflows/review-reusable.yml), dogfooding the Phase-4 "subscribe"
# path. The reusable holds the image pin + env plumbing; this file holds only
# the triggers, the actor gate, and gadfly's specific swarm config.
# #
# Runs the published Gadfly image (pinned to an immutable :sha- tag — act_runner # Advisory only — never blocks a merge. Fleet: 6 cloud + Claude Code
# caches :latest, and this build is what carries foreman provider-type support) # (sonnet, opus, opus:max) competitors.
# 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) name: Adversarial Review (Gadfly)
@@ -33,57 +32,21 @@ jobs:
review: review:
# Security: only trusted users may trigger a secret-bearing run via a PR # Security: only trusted users may trigger a secret-bearing run via a PR
# comment (pull_request + workflow_dispatch are already trusted). Mirrors # comment (pull_request + workflow_dispatch are already trusted). Mirrors
# GADFLY_ALLOWED_USERS, the in-container belt-and-suspenders check. # the allowed_users input below, the in-container belt-and-suspenders check.
if: >- if: >-
github.event_name != 'issue_comment' github.event_name != 'issue_comment'
|| (github.event.issue.pull_request || (github.event.issue.pull_request
&& (github.actor == 'steve' && (github.actor == 'steve'
|| github.actor == 'fizi' || github.actor == 'fizi'
|| github.actor == 'dazed')) || github.actor == 'dazed'))
runs-on: ubuntu-latest uses: ./.gitea/workflows/review-reusable.yml
# Fleet (6 cloud + claude-code/sonnet, all concurrent) reviewing every PR secrets: inherit
# with the 3-lens suite. All cloud now, so runs are fast. with:
timeout-minutes: 90 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,claude-code/opus:max"
steps: specialists: "security,correctness,error-handling"
- uses: docker://gitea.stevedudenhoeffer.com/steve/gadfly:sha-c342bdb provider_concurrency: "ollama-cloud=3,claude-code=2"
env: provider_lens_concurrency: "ollama-cloud=3"
GITEA_API: ${{ github.server_url }}/api/v1/repos/${{ github.repository }} timeout_secs: "600"
GITEA_TOKEN: ${{ secrets.GITEA_TOKEN }} max_steps: "14"
OLLAMA_CLOUD_API_KEY: ${{ secrets.OLLAMA_CLOUD_API_KEY }} allowed_users: "steve,fizi,dazed"
# Claude Code engine auth (for the claude-code/* entries in GADFLY_MODELS timeout_minutes: 90
# 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, and opus
# with max extended thinking) — 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,claude-code/opus:max"
# 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 }}
+93
View File
@@ -0,0 +1,93 @@
# Gadfly — REUSABLE adversarial-review workflow (Gitea `workflow_call`).
#
# Centralizes the ~90-line consumer stub so a repo can subscribe to Gadfly with
# a tiny caller. A consumer workflow does:
#
# jobs:
# review:
# if: ... # actor gate for the comment trigger
# uses: steve/gadfly/.gitea/workflows/review-reusable.yml@main
# secrets: inherit # passes OLLAMA_CLOUD_API_KEY etc. through
# with: { models: "...", allowed_users: "..." } # all optional
#
# Inputs are all optional and default to "" — an empty env value makes the
# image/entrypoint use its own built-in default, so the caller only sets what it
# wants to override. Secrets come via `secrets: inherit` (verified working on
# this Gitea's act_runner); an undefined secret resolves to empty, so optional
# ones (Claude Code token, foreman endpoints, findings store) are harmless when
# the consumer hasn't set them.
#
# Advisory only — never blocks a merge. The image is pinned to an immutable
# :sha- tag here (act_runner caches :latest); bump it per Gadfly release.
name: Gadfly review (reusable)
on:
workflow_call:
inputs:
models: { type: string, default: "" } # GADFLY_MODELS (csv)
specialists: { type: string, default: "" } # GADFLY_SPECIALISTS
provider: { type: string, default: "" } # GADFLY_PROVIDER
base_url: { type: string, default: "" } # GADFLY_BASE_URL
provider_concurrency: { type: string, default: "" } # GADFLY_PROVIDER_CONCURRENCY
provider_lens_concurrency: { type: string, default: "" } # GADFLY_PROVIDER_LENS_CONCURRENCY
timeout_secs: { type: string, default: "" } # GADFLY_TIMEOUT_SECS (per lens)
max_steps: { type: string, default: "" } # GADFLY_MAX_STEPS
worker_model: { type: string, default: "" } # GADFLY_WORKER_MODEL
allowed_users: { type: string, default: "" } # GADFLY_ALLOWED_USERS
trigger_phrase: { type: string, default: "" } # GADFLY_TRIGGER_PHRASE
# Job wall-clock cap. 45 > 30 as a default: a multi-model swarm or a slow
# lens (e.g. claude-code with extended thinking) can exceed 30 minutes.
timeout_minutes: { type: number, default: 45 }
# The reusable job posts the review comment, so it needs issues/PR write. Gitea
# caps these by the caller's granted permissions; declaring them here is explicit.
permissions:
contents: read
issues: write
pull-requests: write
jobs:
review:
runs-on: ubuntu-latest
timeout-minutes: ${{ inputs.timeout_minutes }}
steps:
- uses: docker://gitea.stevedudenhoeffer.com/steve/gadfly:sha-c342bdb
env:
# --- event context (from the CALLER's github.*) -------------------
GITEA_API: ${{ github.server_url }}/api/v1/repos/${{ github.repository }}
GITEA_TOKEN: ${{ secrets.GITEA_TOKEN }}
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 }}
# --- provider auth (via secrets: inherit; empty if consumer unset) -
OLLAMA_CLOUD_API_KEY: ${{ secrets.OLLAMA_CLOUD_API_KEY }}
OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
GOOGLE_API_KEY: ${{ secrets.GOOGLE_API_KEY }}
GADFLY_API_KEY: ${{ secrets.GADFLY_API_KEY }}
CLAUDE_CODE_OAUTH_TOKEN: ${{ secrets.CLAUDE_CODE_OAUTH_TOKEN }}
# Common named foreman/LAN endpoints (optional). Consumers with other
# GADFLY_ENDPOINT_<NAME>s need the full stub (examples/), since a
# reusable workflow can't enumerate arbitrary names.
GADFLY_ENDPOINT_M1: ${{ secrets.GADFLY_ENDPOINT_M1 }}
GADFLY_ENDPOINT_M5: ${{ secrets.GADFLY_ENDPOINT_M5 }}
# --- findings telemetry (optional) --------------------------------
GADFLY_FINDINGS_URL: ${{ secrets.GADFLY_FINDINGS_URL }}
GADFLY_FINDINGS_TOKEN: ${{ secrets.GADFLY_FINDINGS_TOKEN }}
# --- config (from inputs; empty => image default) -----------------
GADFLY_MODELS: ${{ inputs.models }}
GADFLY_SPECIALISTS: ${{ inputs.specialists }}
GADFLY_PROVIDER: ${{ inputs.provider }}
GADFLY_BASE_URL: ${{ inputs.base_url }}
GADFLY_PROVIDER_CONCURRENCY: ${{ inputs.provider_concurrency }}
GADFLY_PROVIDER_LENS_CONCURRENCY: ${{ inputs.provider_lens_concurrency }}
GADFLY_TIMEOUT_SECS: ${{ inputs.timeout_secs }}
GADFLY_MAX_STEPS: ${{ inputs.max_steps }}
GADFLY_WORKER_MODEL: ${{ inputs.worker_model }}
GADFLY_ALLOWED_USERS: ${{ inputs.allowed_users }}
GADFLY_TRIGGER_PHRASE: ${{ inputs.trigger_phrase }}
+3
View File
@@ -46,6 +46,9 @@ entrypoint.sh container brains: trigger gating, PR clone, model loop (t
used to live in workflow YAML) used to live in workflow YAML)
Dockerfile multi-stage; private-module creds via BuildKit secrets never reach the final image Dockerfile multi-stage; private-module creds via BuildKit secrets never reach the final image
.gitea/workflows/build-image.yml push main → :latest; tag v* → :<tag>+:latest; PR → build-only .gitea/workflows/build-image.yml push main → :latest; tag v* → :<tag>+:latest; PR → build-only
.gitea/workflows/review-reusable.yml reusable (workflow_call) review job; consumers subscribe with
an ~8-line caller + `secrets: inherit` (Phase 4). gadfly's own
adversarial-review.yml is a thin caller of it (dogfoods the path).
examples/ copy-paste consumer stub workflows for different providers examples/ copy-paste consumer stub workflows for different providers
``` ```
+6 -3
View File
@@ -37,9 +37,12 @@ Gadfly ships as a container image, so consuming repos don't build anything — t
it. Drop one file in your repo and set a couple of secrets/vars: it. Drop one file in your repo and set a couple of secrets/vars:
1. Copy a stub from [`examples/`](examples/) to `.gitea/workflows/adversarial-review.yml` in 1. Copy a stub from [`examples/`](examples/) to `.gitea/workflows/adversarial-review.yml` in
your repo — [`adversarial-review.yml`](examples/adversarial-review.yml) for the Ollama Cloud your repo. Two flavors: the slim [`reusable.yml`](examples/reusable.yml) — a tiny caller of
default, or a provider-specific one (local Ollama, OpenAI-compatible, endpoint aliases). See Gadfly's **reusable workflow** (`uses: steve/gadfly/.gitea/workflows/review-reusable.yml@…`
the [examples index](examples/README.md). + `secrets: inherit`), best when you take the defaults — or the full self-contained
[`adversarial-review.yml`](examples/adversarial-review.yml) (Ollama Cloud default, with inline
notes for every provider / local Ollama / OpenAI-compatible / endpoint aliases). See the
[examples index](examples/README.md).
2. Add repo config: 2. Add repo config:
- **secret** `OLLAMA_CLOUD_API_KEY` — your [Ollama Cloud](https://ollama.com) key (empty - **secret** `OLLAMA_CLOUD_API_KEY` — your [Ollama Cloud](https://ollama.com) key (empty
⇒ Gadfly posts a harmless "not configured" notice instead of reviewing). *Not needed if ⇒ Gadfly posts a harmless "not configured" notice instead of reviewing). *Not needed if
+2 -1
View File
@@ -6,7 +6,8 @@ set the secrets/vars it references. Gadfly is advisory only — it never blocks
| File | Backend | Needs | | File | Backend | Needs |
|------|---------|-------| |------|---------|-------|
| [`adversarial-review.yml`](adversarial-review.yml) | **Ollama Cloud** (default) + inline notes for every provider | secret `OLLAMA_CLOUD_API_KEY` | | [`reusable.yml`](reusable.yml) | **slimmest stub** — calls Gadfly's reusable workflow (`secrets: inherit`); take the defaults or override a few inputs | secret `OLLAMA_CLOUD_API_KEY` |
| [`adversarial-review.yml`](adversarial-review.yml) | **Ollama Cloud** (default) + inline notes for every provider; full self-contained stub | secret `OLLAMA_CLOUD_API_KEY` |
| [`local-ollama.yml`](local-ollama.yml) | a **local/LAN Ollama** daemon | nothing (or `GADFLY_BASE_URL` for a remote host) | | [`local-ollama.yml`](local-ollama.yml) | a **local/LAN Ollama** daemon | nothing (or `GADFLY_BASE_URL` for a remote host) |
| [`openai-compatible.yml`](openai-compatible.yml) | any **OpenAI-compatible** endpoint (local Ollama `/v1`, gateway, vLLM, OpenRouter…) | `GADFLY_BASE_URL` (+ a key for most gateways) | | [`openai-compatible.yml`](openai-compatible.yml) | any **OpenAI-compatible** endpoint (local Ollama `/v1`, gateway, vLLM, OpenRouter…) | `GADFLY_BASE_URL` (+ a key for most gateways) |
| [`endpoint-aliases.yml`](endpoint-aliases.yml) | **several named backends** at once (one comment each) | repo vars `GADFLY_ENDPOINT_<NAME>` | | [`endpoint-aliases.yml`](endpoint-aliases.yml) | **several named backends** at once (one comment each) | repo vars `GADFLY_ENDPOINT_<NAME>` |
+50
View File
@@ -0,0 +1,50 @@
# Gadfly — SLIM consumer stub via the reusable workflow.
# Copy to .gitea/workflows/adversarial-review.yml in your repo.
#
# This is the shortest way to subscribe: it calls Gadfly's centralized reusable
# workflow, which holds the image pin + all the env plumbing. You only declare
# the triggers, the comment-trigger actor gate, and any overrides you want.
#
# Needs: secret OLLAMA_CLOUD_API_KEY (the default Ollama Cloud provider).
# `secrets: inherit` passes your repo/org/user secrets through to the reusable
# workflow (GITEA_TOKEN is automatic). Pin @<ref> to a Gadfly tag/branch.
#
# Prefer this when you're happy with the defaults. For custom named endpoints
# (GADFLY_ENDPOINT_<NAME>) or a provider the reusable doesn't map, use the full
# stub in adversarial-review.yml instead.
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:
# Only let your maintainers re-trigger via a PR comment (keep in sync with
# the allowed_users override below).
if: >-
github.event_name != 'issue_comment'
|| (github.event.issue.pull_request && github.actor == 'your-username')
# Pin @<ref> to a Gadfly release tag for stability (@main tracks latest).
uses: steve/gadfly/.gitea/workflows/review-reusable.yml@main
secrets: inherit
with:
# All optional — omit to take Gadfly's defaults. Examples:
# models: "qwen3-coder:480b-cloud,gpt-oss:120b-cloud"
# specialists: "security,correctness,error-handling"
allowed_users: "your-username"