/ag-brain-release
Release automation for Agent Brain packages.
> /plugin marketplace add SpillwaveSolutions/agent-brain > /plugin install agent-brain@agent-brain-marketplace
How it fires
How this command gets triggered: by you, by Claude, or both.
- Fires itselfClaude auto-loads it when your prompt matches the work.
- You can call itInvoke it directly when you want it.
- Slash command
/ag-brain-release
Context preview
What this command does when you run it.
Release automation for Agent Brain packages.
Command definition
ag-brain-release.mdcontext: fork
agent: release_agent
ag-brain-release
Release automation for Agent Brain packages.
Arguments
- `<bump>` (required): Version bump type
- `major` - Breaking changes (1.2.0 → 2.0.0)
- `minor` - New features (1.2.0 → 1.3.0)
- `patch` - Bug fixes (1.2.0 → 1.2.1)
- `--dry-run` (optional): Preview changes without executing
Examples
/ag-brain-release patch
/ag-brain-release minor
/ag-brain-release major
/ag-brain-release minor --dry-run
Task
Execute a versioned release with these steps:
Pre-Release Checks (MUST PASS)
1. Working directory is clean (`git status --porcelain` empty) 2. On `main` branch 3. Synced with remote origin/main 4. CLI dependency points to PyPI (not path) - flip if needed 5. CHANGELOG entry exists for the new version. Calculate the new version from current + bump type, then verify `docs/CHANGELOG.md` contains a `## [X.Y.Z]` heading matching it. If missing, abort with: "Add a `## [X.Y.Z] - YYYY-MM-DD` section to docs/CHANGELOG.md and commit it before re-running. See the existing [10.0.0] entry for the expected Keep-a-Changelog format."
> **Note**: `task before-push` is wrapped by an automatic lock-drift guard (`scripts/before_push_lock_guard.sh`, see issue #174). The guard snapshots clean `agent-brain-{server,cli}/poetry.lock` files at task entry and reverts any in-task churn on exit. The release flow inherits this protection — no separate manual `git checkout HEAD -- poetry.lock` step is required after `task before-push`.
Release Steps
1. Calculate new version from current + bump type (also used by pre-release check #5) 2. Flip CLI dependency to PyPI if path-based 3. Update version in these files (the CI script `scripts/check_plugin_manifest_versions.sh` is the checklist):
- `agent-brain-server/pyproject.toml`
- `agent-brain-server/agent_brain_server/__init__.py`
- `agent-brain-cli/pyproject.toml`
- `agent-brain-cli/agent_brain_cli/__init__.py`
- `agent-brain-uds/pyproject.toml`
- `agent-brain-uds/agent_brain_uds/__init__.py`
- `agent-brain-mcp/pyproject.toml`
- `agent-brain-mcp/agent_brain_mcp/__init__.py`
- `agent-brain-plugin/.claude-plugin/plugin.json`
- `.claude-plugin/marketplace.json` (`plugins[0].version`)
- `agent-brain-plugin/plugin.json` (Agent Plugins 1.0)
- `agent-brain-plugin/.codex-plugin/plugin.json`
- `agent-brain-plugin/.cursor-plugin/plugin.json`
- `agent-brain-plugin/.grok-plugin/marketplace.json` (top-level + `plugins[0].version`)
4. Commit: `chore(release): bump version to X.Y.Z` 5. Tag: `vX.Y.Z` 6. Push branch and tag 7. Create GitHub release (triggers PyPI publish)
Dry-Run Mode
If `--dry-run`, report what WOULD happen without executing.
Expected Result
Report:
- Pre-check status (clean/branch/sync/dep)
- Version calculation (current → new)
- Files updated
- Git operations performed
- GitHub release URL
- PyPI package URLs (available in ~5 minutes)
Read more
context: fork agent: release_agent
ag-brain-release
Release automation for Agent Brain packages.
Arguments
- `<bump>` (required): Version bump type
- `major` - Breaking changes (1.2.0 → 2.0.0)
- `minor` - New features (1.2.0 → 1.3.0)
- `patch` - Bug fixes (1.2.0 → 1.2.1)
- `--dry-run` (optional): Preview changes without executing
Examples
/ag-brain-release patch /ag-brain-release minor /ag-brain-release major /ag-brain-release minor --dry-run
Task
Execute a versioned release with these steps:
Pre-Release Checks (MUST PASS)
1. Working directory is clean (`git status --porcelain` empty) 2. On `main` branch 3. Synced with remote origin/main 4. CLI dependency points to PyPI (not path) - flip if needed 5. CHANGELOG entry exists for the new version. Calculate the new version from current + bump type, then verify `docs/CHANGELOG.md` contains a `## [X.Y.Z]` heading matching it. If missing, abort with: "Add a `## [X.Y.Z] - YYYY-MM-DD` section to docs/CHANGELOG.md and commit it before re-running. See the existing [10.0.0] entry for the expected Keep-a-Changelog format."
> **Note**: `task before-push` is wrapped by an automatic lock-drift guard (`scripts/before_push_lock_guard.sh`, see issue #174). The guard snapshots clean `agent-brain-{server,cli}/poetry.lock` files at task entry and reverts any in-task churn on exit. The release flow inherits this protection — no separate manual `git checkout HEAD -- poetry.lock` step is required after `task before-push`.
Release Steps
1. Calculate new version from current + bump type (also used by pre-release check #5) 2. Flip CLI dependency to PyPI if path-based 3. Update version in these files (the CI script `scripts/check_plugin_manifest_versions.sh` is the checklist):
- `agent-brain-server/pyproject.toml`
- `agent-brain-server/agent_brain_server/__init__.py`
- `agent-brain-cli/pyproject.toml`
- `agent-brain-cli/agent_brain_cli/__init__.py`
- `agent-brain-uds/pyproject.toml`
- `agent-brain-uds/agent_brain_uds/__init__.py`
- `agent-brain-mcp/pyproject.toml`
- `agent-brain-mcp/agent_brain_mcp/__init__.py`
- `agent-brain-plugin/.claude-plugin/plugin.json`
- `.claude-plugin/marketplace.json` (`plugins[0].version`)
- `agent-brain-plugin/plugin.json` (Agent Plugins 1.0)
- `agent-brain-plugin/.codex-plugin/plugin.json`
- `agent-brain-plugin/.cursor-plugin/plugin.json`
- `agent-brain-plugin/.grok-plugin/marketplace.json` (top-level + `plugins[0].version`)
4. Commit: `chore(release): bump version to X.Y.Z` 5. Tag: `vX.Y.Z` 6. Push branch and tag 7. Create GitHub release (triggers PyPI publish)
Dry-Run Mode
If `--dry-run`, report what WOULD happen without executing.
Expected Result
Report:
- Pre-check status (clean/branch/sync/dep)
- Version calculation (current → new)
- Files updated
- Git operations performed
- GitHub release URL
- PyPI package URLs (available in ~5 minutes)
A RAG-based (Retrieval-Augmented Generation) document indexing and semantic search system for AI agents and applications. Agent Brain enables intelligent querying of documentation and source code using natural language.
Other commands on agent-brain.
ag-install-local
Local build-and-install of Agent Brain server + CLI + plugin for rapid testing.
ag-remove-local-instal…
Remove all local Agent Brain installations, caches, and plugin artifacts.
implement-python
Senior Python engineer implementation command. Launches a background agent with full Python development capabilities and access…
speckit.analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task…
speckit.checklist
Generate a custom checklist for the current feature based on user requirements.

