mirror of
https://github.com/block/buzz.git
synced 2026-10-27 12:40:30 +00:00
## Summary Replacing `BUZZ_ACP_AGENT_COMMAND` with a sandbox launcher hides the real adapter from Buzz: for example, Goose loses its default `acp` argument. Add optional `BUZZ_ACP_LAUNCH_PREFIX`, a JSON argument array applied at the shared subprocess boundary, after normal worker configuration. On Unix, every spawn runs `prefix... worker args...` without shell interpolation. Worker identity, environment setup, stdio and existing process cleanup are preserved. Invalid or unavailable prefixes fail launch without falling back to the worker; unset keeps direct launch on every platform. Configured prefixes fail closed on non-Unix because per-worker process-tree cleanup is unavailable. Desktop reserves the key against saved user-environment overrides, and spawn failures name the executable. ### Related issue Prerequisite for [Buzz-App #415](https://github.com/block/buzz-app/pull/415). One commit directly on `main`, independent of #7942–#7944. No duplicate launch-prefix PR found. Policy enforcement, verified launcher staging, protected paths and the supporting runtime pin remain Buzz-App/plugin work. ### Testing - Original head `9ae9d82f8`: complete ACP suite (990 passed, three existing ignored), shipped local-task test, real-Goose probe, and full repository-wide `just ci` passed. - Review fixes: five launch tests and 44 Desktop environment-filter tests pass locally, covering the real production platform gate, save rejection, both merge layers, case variants, and executable error context. - Subprocess tests cover direct/wrapped argument defaults, Pi skills, Hermes/Codex environment, adapter identity, repeated spawns and invalid/missing prefix refusal. - Shipped `buzz-acp run` entrypoint completes a task through the wrapper and deterministic ACP peer. - Real Goose 1.52.0: direct and wrapped `auth-methods` return identical results; the wrapper receives `goose acp` with initially empty worker args. - Independent agent review: no blocking findings. - Required push checks passed on the updated head. The existing native Windows CI job now explicitly runs the platform-contract tests. Its result is pending; the non-Unix refusal branch was not exercised on this macOS host. Repeated-spawn coverage is not a full lazy-pool wake/crash test. Protected Goose in the Buzz-App UI and human acceptance remain pending. To try the hook after building `buzz-acp`, use an installed Goose path: ```sh BUZZ_ACP_LAUNCH_PREFIX='["/usr/bin/env"]' target/debug/buzz-acp auth-methods \ --agent-command /absolute/path/to/goose --agent-args '' --json ``` Expect the same authentication methods as without the prefix. Changing the prefix to `["/missing-launcher"]` must fail. Draft pending human testing. Signed-off-by: Alex Rosenzweig <arosenzweig@squareup.com>
223 lines
9.6 KiB
YAML
223 lines
9.6 KiB
YAML
name: CI / Rust
|
|
on:
|
|
workflow_call:
|
|
inputs:
|
|
rust:
|
|
required: true
|
|
type: boolean
|
|
desktop_rust:
|
|
required: true
|
|
type: boolean
|
|
lane:
|
|
required: true
|
|
type: string
|
|
outputs:
|
|
rust_lint_result:
|
|
value: ${{ jobs.results.outputs.rust_lint_result }}
|
|
unit_tests_result:
|
|
value: ${{ jobs.results.outputs.unit_tests_result }}
|
|
windows_rust_result:
|
|
value: ${{ jobs.results.outputs.windows_rust_result }}
|
|
|
|
env:
|
|
CARGO_TERM_COLOR: always
|
|
BUZZ_TEST_POSTGRES_PASSWORD: buzz_dev
|
|
PLAYWRIGHT_BROWSERS_PATH: ${{ github.workspace }}/.cache/ms-playwright
|
|
|
|
jobs:
|
|
rust-lint:
|
|
name: Rust Lint
|
|
runs-on: ubuntu-latest
|
|
timeout-minutes: 30
|
|
if: inputs.lane == 'required' && (github.event_name == 'push' || inputs.rust || inputs.desktop_rust)
|
|
permissions:
|
|
contents: read
|
|
steps:
|
|
- uses: actions/checkout@df4cb1c069e1874edd31b4311f1884172cec0e10 # v6.0.3
|
|
- uses: cashapp/activate-hermit@cea9af7913204a965fd488637a8d1811bba2e616 # v1
|
|
- uses: Swatinem/rust-cache@e18b497796c12c097a38f9edb9d0641fb99eee32
|
|
with:
|
|
save-if: ${{ github.event_name != 'pull_request' }}
|
|
- name: Format check
|
|
run: just fmt-check
|
|
- name: Desktop Tauri format check
|
|
run: just desktop-tauri-fmt-check
|
|
- name: Clippy
|
|
run: just clippy
|
|
|
|
unit-tests:
|
|
name: Unit Tests
|
|
runs-on: ubuntu-latest
|
|
timeout-minutes: 30
|
|
if: inputs.lane == 'required' && (github.event_name == 'push' || inputs.rust)
|
|
permissions:
|
|
contents: read
|
|
steps:
|
|
- uses: actions/checkout@df4cb1c069e1874edd31b4311f1884172cec0e10 # v6.0.3
|
|
- uses: cashapp/activate-hermit@cea9af7913204a965fd488637a8d1811bba2e616 # v1
|
|
- uses: rui314/setup-mold@7e4f20ad28a2e8ca6fd0892ccf72e2abb706b9c3 # v1
|
|
- uses: Swatinem/rust-cache@e18b497796c12c097a38f9edb9d0641fb99eee32
|
|
with:
|
|
key: sherpa-cache-v1
|
|
save-if: ${{ github.event_name != 'pull_request' }}
|
|
- name: Install cargo-nextest
|
|
uses: taiki-e/install-action@0fd46367812ee04360509b4169d9f659d6892bb2 # v2.79.15
|
|
with:
|
|
tool: cargo-nextest@0.9.136
|
|
- name: Unit tests
|
|
run: just test-unit
|
|
|
|
server-cross-compile:
|
|
name: Server Cross-Compile
|
|
runs-on: ubuntu-latest
|
|
timeout-minutes: 30
|
|
if: inputs.lane == 'cross-compile' && (github.event_name == 'push' || inputs.rust)
|
|
permissions:
|
|
contents: read
|
|
strategy:
|
|
fail-fast: false
|
|
matrix:
|
|
target:
|
|
- x86_64-unknown-linux-musl
|
|
- aarch64-unknown-linux-musl
|
|
steps:
|
|
- uses: actions/checkout@df4cb1c069e1874edd31b4311f1884172cec0e10 # v6.0.3
|
|
- uses: cashapp/activate-hermit@cea9af7913204a965fd488637a8d1811bba2e616 # v1
|
|
- uses: Swatinem/rust-cache@e18b497796c12c097a38f9edb9d0641fb99eee32
|
|
with:
|
|
key: cross-${{ matrix.target }}
|
|
save-if: ${{ github.event_name != 'pull_request' }}
|
|
- name: Install cross
|
|
uses: taiki-e/install-action@0fd46367812ee04360509b4169d9f659d6892bb2 # v2.79.15
|
|
with:
|
|
tool: cross@0.2.5
|
|
- name: Build server binaries
|
|
env:
|
|
TARGET: ${{ matrix.target }}
|
|
# PRs: compile + build-script gate only (no codegen/link). Main: full link gate.
|
|
CARGO_CMD: ${{ github.event_name == 'pull_request' && 'check' || 'build' }}
|
|
run: |
|
|
cross "$CARGO_CMD" --release --target "$TARGET" \
|
|
-p buzz-relay \
|
|
-p buzz-acp \
|
|
-p buzz-agent \
|
|
-p buzz-dev-mcp \
|
|
-p git-credential-nostr \
|
|
-p git-sign-nostr
|
|
|
|
windows-rust:
|
|
name: Windows Rust (x86_64-pc-windows-msvc)
|
|
runs-on: windows-latest
|
|
# Windows runners are slow and this compiles the workspace + Tauri crate
|
|
# cold across four steps; budget generously.
|
|
timeout-minutes: 45
|
|
if: inputs.lane == 'required' && (github.event_name == 'push' || inputs.rust || inputs.desktop_rust)
|
|
permissions:
|
|
contents: read
|
|
env:
|
|
TARGET: x86_64-pc-windows-msvc
|
|
steps:
|
|
- uses: actions/checkout@df4cb1c069e1874edd31b4311f1884172cec0e10 # v6.0.3
|
|
# MSVC needs windows.h (aws-lc-sys et al.), so this runs on a real Windows
|
|
# runner — hermit, used by the Linux jobs, does not provide MSVC. The
|
|
# toolchain (1.95.0 + clippy via profile = default) comes from the
|
|
# repo-root rust-toolchain.toml, which the runner's preinstalled rustup
|
|
# honors on demand; the host triple already is x86_64-pc-windows-msvc.
|
|
- uses: Swatinem/rust-cache@e18b497796c12c097a38f9edb9d0641fb99eee32
|
|
with:
|
|
workspaces: |
|
|
.
|
|
desktop/src-tauri
|
|
key: windows-msvc
|
|
save-if: ${{ github.event_name != 'pull_request' }}
|
|
# Tauri validates externalBin at compile time, so the Tauri-crate steps
|
|
# below fail without these stubs. Mirrors scripts/bundle-sidecars.sh's
|
|
# Windows naming (binaries/<bin>-<triple>.exe); empty files suffice for a
|
|
# type-check since nothing executes them.
|
|
- name: Create sidecar placeholders
|
|
shell: bash
|
|
run: |
|
|
mkdir -p desktop/src-tauri/binaries
|
|
for bin in buzz-acp buzz-agent buzz-dev-mcp git-credential-nostr buzz; do
|
|
touch "desktop/src-tauri/binaries/${bin}-${TARGET}.exe"
|
|
done
|
|
- name: Clippy (workspace)
|
|
run: cargo clippy --workspace --all-targets --target $env:TARGET -- -D warnings
|
|
- name: Check (workspace)
|
|
run: cargo check --workspace --all-targets --target $env:TARGET
|
|
- name: Test (buzz-dev-mcp)
|
|
# The Windows-only bash resolver lives in buzz-dev-mcp; its unit tests
|
|
# only gate if this crate is tested ON Windows.
|
|
# Serial: windows_resolver_tests mutate process-global env
|
|
# (BUZZ_SHELL/GIT_BASH/SystemRoot) that SharedState::new reads.
|
|
run: cargo test -p buzz-dev-mcp --target $env:TARGET -- --test-threads=1
|
|
- name: Test (ACP launch-prefix platform contract)
|
|
# Execute the non-Unix refusal branch; a cross-compile cannot test it.
|
|
run: cargo test -p buzz-acp --target $env:TARGET --lib acp::launch::platform_tests
|
|
- name: Test (buzz-agent auth coordinator)
|
|
# The auth coordinator single-flights on an OS advisory lock, which is
|
|
# LockFileEx on Windows; this integration suite drives real second
|
|
# processes on the same lock file, so it only exercises the Windows
|
|
# lock runtime if it runs ON Windows. Every other job compiles it but
|
|
# never executes it. Tests exercised on Windows: lock serialization
|
|
# (two coordinators race for the same key), cooldown sidecar sharing
|
|
# across processes, attempt-sidecar adoption (UserInitiated waiter
|
|
# adopts a predecessor's denial), and the in-process single-flight for
|
|
# same-key coalescing. Tests that are UNIX-ONLY and NOT executed here:
|
|
# crash-release (flock drop on SIGKILL, guarded by #[cfg(unix)]) and
|
|
# cross-process cache success/race (on-disk token handoff, also
|
|
# #[cfg(unix)]).
|
|
run: cargo test -p buzz-agent --target $env:TARGET --test databricks_auth_coordinator
|
|
# Smoke-test the new host-prereq contract: Git for Windows (which provides
|
|
# bash) is available on the runner, a shell command round-trips, and bash
|
|
# does NOT resolve from System32 (so WSL's launcher is never picked up).
|
|
# windows-latest runners have Git for Windows pre-installed; the unit tests
|
|
# above exercise the MCP resolver itself. This step verifies the host env.
|
|
- name: Smoke-test host Git Bash prereq (host env check)
|
|
shell: bash
|
|
run: |
|
|
set -euo pipefail
|
|
# Git for Windows ships bash.exe under its bin/ directory; confirm it
|
|
# resolves from the standard location the runtime resolver probes first.
|
|
bash_path=$(command -v bash 2>/dev/null || true)
|
|
[[ -n "$bash_path" ]] || { echo "ERROR: bash not found on PATH — host Git for Windows missing" >&2; exit 1; }
|
|
echo "Resolved bash: $bash_path"
|
|
[[ "$bash_path" != *System32* ]] || { echo "ERROR: resolved bash is WSL's System32 launcher" >&2; exit 1; }
|
|
|
|
# Run a basic pipeline through the resolved bash (same invocation the
|
|
# agent uses: bash -c '...').
|
|
out=$(bash -c 'echo hello | tr a-z A-Z')
|
|
[[ "$out" == "HELLO" ]] || { echo "bash pipeline failed: got '$out'" >&2; exit 1; }
|
|
|
|
# Confirm git itself works — agents run git commands frequently.
|
|
git --version
|
|
repo=$(mktemp -d)
|
|
cd "$repo"
|
|
git init -q
|
|
git -c user.name=ci -c user.email=ci@example.com commit -q --allow-empty -m smoke
|
|
git log -1 --format=%s | grep -qx smoke
|
|
echo "Host bash resolved and functional; git commit round-trip passed"
|
|
- name: Check (Tauri crate)
|
|
run: cargo check --manifest-path desktop/src-tauri/Cargo.toml --workspace --all-targets --target $env:TARGET
|
|
env:
|
|
CMAKE_POLICY_VERSION_MINIMUM: "3.5"
|
|
- name: Test (Tauri crate)
|
|
run: cargo test --manifest-path desktop/src-tauri/Cargo.toml --target $env:TARGET
|
|
env:
|
|
CMAKE_POLICY_VERSION_MINIMUM: "3.5"
|
|
|
|
|
|
results:
|
|
name: Results
|
|
if: ${{ always() }}
|
|
needs: [rust-lint, unit-tests, server-cross-compile, windows-rust]
|
|
runs-on: ubuntu-latest
|
|
timeout-minutes: 5
|
|
permissions: {}
|
|
outputs:
|
|
rust_lint_result: ${{ needs.rust-lint.result }}
|
|
unit_tests_result: ${{ needs.unit-tests.result }}
|
|
windows_rust_result: ${{ needs.windows-rust.result }}
|
|
steps:
|
|
- run: echo "Captured Rust job results"
|