refactor: SandboxAgent + SDK Shell/Filesystem; agent-browser CLI; nuke FastAPI sidecar

Combined commits 2+3 of the migration plan because the FastAPI sidecar
removal in commit 2 broke ``browser_action`` (which lived in the
sidecar); they have to land together.

Sandbox tool layer (commit 2 piece):
- ``build_strix_agent`` now returns a ``SandboxAgent`` with
  ``capabilities=[Filesystem(), Shell()]``. The SDK runtime binds the
  capabilities to the live sandbox session per-run; agents get
  ``exec_command``, ``write_stdin``, ``apply_patch``, ``view_image``
  function tools auto-merged into their tool list. Plain ``Agent``
  short-circuits capability binding (``agents/sandbox/runtime.py:190``).
- Drop ``Compaction`` from the default capability set — it's
  OpenAI-Responses-API-only and useless for our litellm-routed
  Anthropic setup.
- Delete the entire custom in-container tool layer:
  - ``strix/tools/terminal/`` (5 files, 748 LoC libtmux)
  - ``strix/tools/file_edit/`` (3 files, 276 LoC)
  - ``strix/tools/python/`` (5 files, 459 LoC)
  - ``strix/runtime/tool_server.py`` (163 LoC FastAPI sidecar)
  - ``strix/tools/_sandbox_dispatch.py`` (117 LoC)
  - ``strix/tools/registry.py`` (109 LoC)
  - ``strix/tools/context.py`` (12 LoC)
- Drop the corresponding TUI renderers (``terminal_renderer.py``,
  ``file_edit_renderer.py``, ``python_renderer.py``) and update
  ``interface/tool_components/__init__.py``.

Browser → agent-browser CLI (commit 3 piece):
- Install ``agent-browser@0.26.0`` globally in the Dockerfile right
  after the existing ``npm install -g`` block. Run
  ``agent-browser install --with-deps`` (apt, root) and
  ``agent-browser install`` (Chrome download, pentester) +
  ``agent-browser doctor --offline --quick`` smoke test.
- Drop the explicit Playwright system-deps apt list (replaced by
  ``--with-deps``) and ``RUN .venv/bin/python -m playwright install
  chromium``.
- Vendor ``agent-browser/skill-data/core/SKILL.md`` →
  ``strix/skills/tooling/agent_browser.md`` (476 lines). Adapt
  frontmatter to Strix format; strip the install/Quickstart and the
  ``agent-browser skills get electron|slack|...`` specialized-skills
  block; add the "Caido proxy is wired via env vars; do not pass
  ``--proxy``" note.
- ``_resolve_skills`` now eagerly loads ``tooling/agent_browser`` for
  every agent (matches the previous unconditional ``browser_action``
  in ``_BASE_TOOLS``).
- Delete ``strix/tools/browser/`` (5 files, 1338 LoC) and the
  ``browser_renderer.py`` TUI render.

Sandbox plumbing:
- Drop ``bearer`` token, ``tool_server_host_port`` resolution + bundle
  keys, ``TOOL_SERVER_TOKEN``/``TOOL_SERVER_PORT``/
  ``STRIX_SANDBOX_EXECUTION_TIMEOUT`` from the manifest env in
  ``session_manager.create_or_reuse``. Caido proxy env vars
  (``http_proxy``, ``https_proxy``, ``ALL_PROXY``) stay; manifest
  applies them to every ``docker exec``-spawned process.
- Drop ``sandbox_token`` and ``tool_server_host_port`` params from
  ``make_agent_context`` and the ``create_agent`` graph tool.
- Drop the tool-server health-check from ``entry.py`` (only Caido's
  ``wait_for_tcp_ready`` remains).
- ``docker-entrypoint.sh``: delete the ~30 line
  ``Starting tool server...`` block (sudo + uvicorn launch + curl
  /health poll). Add ``NO_PROXY=localhost,127.0.0.1`` to
  ``/etc/profile.d/proxy.sh`` and ``/etc/environment`` so the
  agent-browser daemon's CDP traffic on localhost isn't routed
  through Caido.

pyproject.toml:
- ``[project.optional-dependencies] sandbox = []`` (every member of
  the previous list — fastapi, uvicorn, ipython, openhands-aci,
  playwright, libtmux — is gone with the sidecar).
- Drop ``numpydoc.*``, ``IPython.*``, ``openhands_aci.*``,
  ``playwright.*``, ``uvicorn.*``, ``pyte.*``, ``libtmux.*`` from
  the missing-imports module list.
- Drop the per-file ruff ignores for the deleted modules.

Net delta: −5512 LoC. ruff drops to 3 errors (was 21 baseline). mypy
falls to 69 errors over 3 files (was 84 over 8 — the drop comes from
deleting the modules with the worst untyped-import problems).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
0xallam
2026-04-25 14:33:38 -07:00
parent 5449af2456
commit 2c2ab13c8f
40 changed files with 527 additions and 5572 deletions
+1 -27
View File
@@ -49,14 +49,7 @@ dependencies = [
strix = "strix.interface.main:main"
[project.optional-dependencies]
sandbox = [
"fastapi",
"uvicorn",
"ipython>=9.3.0",
"openhands-aci>=0.3.0",
"playwright>=1.48.0",
"libtmux>=0.46.2",
]
sandbox = []
[dependency-groups]
dev = [
@@ -104,20 +97,13 @@ pretty = true
[[tool.mypy.overrides]]
module = [
"litellm.*",
"numpydoc.*",
"rich.*",
"IPython.*",
"openhands_aci.*",
"playwright.*",
"uvicorn.*",
"jinja2.*",
"pydantic_settings.*",
"jwt.*",
"httpx.*",
"gql.*",
"textual.*",
"pyte.*",
"libtmux.*",
"cvss.*",
"scrubadub.*",
"docker.*",
@@ -246,10 +232,6 @@ ignore = [
"strix/tools/todo/tools.py" = ["TC002"]
"strix/tools/thinking/tool.py" = ["TC002"]
"strix/tools/web_search/tool.py" = ["TC002"]
"strix/tools/file_edit/tools.py" = ["TC002"]
"strix/tools/browser/tool.py" = ["TC002"]
"strix/tools/terminal/tool.py" = ["TC002"]
"strix/tools/python/tool.py" = ["TC002"]
"strix/tools/proxy/tools.py" = ["TC002", "PLR0911"]
"strix/tools/agents_graph/tools.py" = ["TC002"]
"strix/agents/factory.py" = ["TC002"]
@@ -272,14 +254,6 @@ ignore = [
"strix/interface/main.py" = ["BLE001", "PLC0415", "PLR0912", "PLR0915"]
"strix/interface/streaming_parser.py" = ["PLC0415"]
"strix/interface/tool_components/agent_message_renderer.py" = ["PLC0415"]
# Each short-circuit error return in the sandbox dispatch is a distinct,
# documented failure mode the model needs to see verbatim.
"strix/tools/_sandbox_dispatch.py" = ["PLR0911"]
# StrixSession + StrixTracingProcessor catch broad Exception
# intentionally: compressor / sanitizer / disk failures must not tear
# down the run. Calls already log at exception level.
"strix/llm/strix_session.py" = ["BLE001"]
"strix/telemetry/strix_processor.py" = ["BLE001"]
[tool.ruff.lint.isort]
force-single-line = false