feat: add release workflow for version tag sync

.github/workflows/release.yml

@@ -0,0 +1,62 @@
+name: Release
+
+on:
+  push:
+    tags:
+      - "v[0-9]+.[0-9]+.[0-9]+*"
+
+permissions:
+  contents: write # required for gh release create and git push (v0 tag)
+
+concurrency:
+  group: release
+  cancel-in-progress: false
+
+jobs:
+  release:
+    runs-on: ubuntu-24.04
+    steps:
+      - uses: actions/checkout@9c091bb21b7c1c1d1991bb908d89e4e9dddfe3e0 # v7.0.0
+        with:
+          fetch-depth: 0
+
+      - name: Create GitHub Release
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          set -euo pipefail
+          TAG="${GITHUB_REF_NAME}"
+
+          if gh release view "${TAG}" >/dev/null 2>&1; then
+            echo "Release ${TAG} already exists, skipping"
+            exit 0
+          fi
+
+          PRERELEASE=""
+          if [[ "${TAG}" == *-* ]]; then
+            PRERELEASE="--prerelease"
+          fi
+
+          gh release create "${TAG}" \
+            --title "${TAG}" \
+            --notes "Synced with [fullsend ${TAG}](https://github.com/fullsend-ai/fullsend/releases/tag/${TAG})" \
+            ${PRERELEASE}
+
+      - name: Move v0 floating tag
+        if: "!contains(github.ref_name, '-')"
+        run: |
+          set -euo pipefail
+          CURRENT=""
+          if git rev-parse v0 >/dev/null 2>&1; then
+            CURRENT=$(git rev-parse v0)
+            if ! git merge-base --is-ancestor "${CURRENT}" "${GITHUB_SHA}"; then
+              echo "::warning::v0 already points at a newer commit, skipping"
+              exit 0
+            fi
+          fi
+          git tag -f v0 "${GITHUB_SHA}"
+          if [[ -n "${CURRENT}" ]]; then
+            git push --force-with-lease="v0:${CURRENT}" origin v0
+          else
+            git push origin v0
+          fi

AGENTS.md

@@ -58,7 +58,27 @@ Determine:
 Use these criteria as checkpoints. If a checkpoint fails, fix the root
 cause — do not weaken the check.
 
-## 6. Skill resolution
+## 6. Versioning and releases
+
+This repository is versioned in lockstep with
+[fullsend](https://github.com/fullsend-ai/fullsend). Version tags are
+not created here directly — they are pushed by fullsend's release
+workflow after GoReleaser succeeds.
+
+**Workflows:**
+
+- `fullsend.yaml` — centrally managed by fullsend for agent event
+  dispatch. Do not modify without coordinating with the fullsend repo.
+- `release.yml` — repo-specific release automation. Triggered by
+  semver tag pushes from fullsend's release workflow. Creates a GitHub
+  Release and moves the `v0` floating tag.
+
+**The `v0` tag** is a floating tag that always points to the latest
+stable (non-prerelease) version. Downstream consumers can reference
+`@v0` to track the latest release. Pre-release tags (`-rc.N`,
+`-alpha.N`, `-beta.N`) do not move `v0`.
+
+## 7. Skill resolution
 
 Skills declared in agent frontmatter `skills:` arrays are resolved at
 runtime from multiple sources in priority order: repo-level

README.md

@@ -0,0 +1,52 @@
+# fullsend agents
+
+First-class agents for the [fullsend](https://github.com/fullsend-ai/fullsend) platform. These agents automate the software development lifecycle on GitHub — from issue triage through code implementation, review, fix, prioritization, and retrospective analysis.
+
+## Agents
+
+| Agent | Description | Trigger |
+|-------|-------------|---------|
+| **Triage** | Assesses issue sufficiency, searches for duplicates, applies control labels | New issues, `/fs-triage` |
+| **Code** | Implements fixes and features following repo conventions | `ready-to-code` label, `/fs-code` |
+| **Review** | Dispatches parallel sub-agents across six review dimensions | PR events, `/fs-review` |
+| **Fix** | Implements targeted fixes from review feedback | Review comments, `/fs-fix` |
+| **Prioritize** | Scores issues using the RICE framework | Schedule, `/fs-prioritize` |
+| **Retro** | Analyzes completed workflows and proposes improvements | PR close, `/fs-retro` |
+
+See [`docs/`](docs/) for detailed documentation on each agent.
+
+## Repository structure
+
+```
+agents/      Agent system prompts (one per agent)
+docs/        User-facing documentation
+harness/     Harness configurations (sandbox image, timeout, scripts, plugins)
+policies/    Sandbox security policies (filesystem, network, binary restrictions)
+env/         Per-agent environment variables
+schemas/     JSON Schema for validating agent structured output
+scripts/     Pre-scripts (input validation) and post-scripts (GitHub mutations)
+skills/      Reusable skill definitions loaded by agents at runtime
+plugins/     Sandbox plugins (e.g. gopls LSP for the code agent)
+common/      Shared configuration (GCP Vertex AI auth)
+```
+
+## Architecture
+
+Agents run inside sandboxed containers with strict filesystem, network, and binary restrictions. Each agent follows a three-phase pipeline:
+
+1. **Pre-script** — runs on the GitHub Actions runner to validate inputs and prepare the environment
+2. **Sandbox** — runs the agent with restricted permissions; the agent writes code and produces structured JSON output
+3. **Post-script** — runs on the runner with elevated permissions to perform GitHub mutations (pushing branches, creating PRs, posting comments, applying labels)
+
+The agent never has direct write access to the repository. All mutations flow through post-scripts.
+
+## Versioning
+
+This repository is versioned in lockstep with [fullsend](https://github.com/fullsend-ai/fullsend). Tags are pushed by fullsend's release workflow after a successful release — they are not created here directly. The `v0` floating tag always points to the latest stable (non-prerelease) version.
+
+## Workflows
+
+| File | Managed by | Purpose |
+|------|-----------|---------|
+| `fullsend.yaml` | fullsend (centrally managed) | Routes GitHub events to agent dispatch workflows |
+| `release.yml` | This repo | Creates GitHub Releases and moves the `v0` tag on version tag push |