Initial commit: hardened DeerFlow factory
Vendored deer-flow upstream (bytedance/deer-flow) plus prompt-injection hardening: - New deerflow.security package: content_delimiter, html_cleaner, sanitizer (8 layers — invisible chars, control chars, symbols, NFC, PUA, tag chars, horizontal whitespace collapse with newline/tab preservation, length cap) - New deerflow.community.searx package: web_search, web_fetch, image_search backed by a private SearX instance, every external string sanitized and wrapped in <<<EXTERNAL_UNTRUSTED_CONTENT>>> delimiters - All native community web providers (ddg_search, tavily, exa, firecrawl, jina_ai, infoquest, image_search) replaced with hard-fail stubs that raise NativeWebToolDisabledError at import time, so a misconfigured tool.use path fails loud rather than silently falling back to unsanitized output - Native client back-doors (jina_client.py, infoquest_client.py) stubbed too - Native-tool tests quarantined under tests/_disabled_native/ (collect_ignore_glob via local conftest.py) - Sanitizer Layer 7 fix: only collapse horizontal whitespace, preserve newlines and tabs so list/table structure survives - Hardened runtime config.yaml references only the searx-backed tools - Factory overlay (backend/) kept in sync with deer-flow tree as a reference / source See HARDENING.md for the full audit trail and verification steps.
This commit is contained in:
197
deer-flow/backend/tests/test_skills_custom_router.py
Normal file
197
deer-flow/backend/tests/test_skills_custom_router.py
Normal file
@@ -0,0 +1,197 @@
|
||||
import json
|
||||
from pathlib import Path
|
||||
from types import SimpleNamespace
|
||||
|
||||
from fastapi import FastAPI
|
||||
from fastapi.testclient import TestClient
|
||||
|
||||
from app.gateway.routers import skills as skills_router
|
||||
from deerflow.skills.manager import get_skill_history_file
|
||||
from deerflow.skills.types import Skill
|
||||
|
||||
|
||||
def _skill_content(name: str, description: str = "Demo skill") -> str:
|
||||
return f"---\nname: {name}\ndescription: {description}\n---\n\n# {name}\n"
|
||||
|
||||
|
||||
async def _async_scan(decision: str, reason: str):
|
||||
from deerflow.skills.security_scanner import ScanResult
|
||||
|
||||
return ScanResult(decision=decision, reason=reason)
|
||||
|
||||
|
||||
def _make_skill(name: str, *, enabled: bool) -> Skill:
|
||||
skill_dir = Path(f"/tmp/{name}")
|
||||
return Skill(
|
||||
name=name,
|
||||
description=f"Description for {name}",
|
||||
license="MIT",
|
||||
skill_dir=skill_dir,
|
||||
skill_file=skill_dir / "SKILL.md",
|
||||
relative_path=Path(name),
|
||||
category="public",
|
||||
enabled=enabled,
|
||||
)
|
||||
|
||||
|
||||
def test_custom_skills_router_lifecycle(monkeypatch, tmp_path):
|
||||
skills_root = tmp_path / "skills"
|
||||
custom_dir = skills_root / "custom" / "demo-skill"
|
||||
custom_dir.mkdir(parents=True, exist_ok=True)
|
||||
(custom_dir / "SKILL.md").write_text(_skill_content("demo-skill"), encoding="utf-8")
|
||||
config = SimpleNamespace(
|
||||
skills=SimpleNamespace(get_skills_path=lambda: skills_root, container_path="/mnt/skills"),
|
||||
skill_evolution=SimpleNamespace(enabled=True, moderation_model_name=None),
|
||||
)
|
||||
monkeypatch.setattr("deerflow.config.get_app_config", lambda: config)
|
||||
monkeypatch.setattr("deerflow.skills.manager.get_app_config", lambda: config)
|
||||
monkeypatch.setattr("app.gateway.routers.skills.scan_skill_content", lambda *args, **kwargs: _async_scan("allow", "ok"))
|
||||
refresh_calls = []
|
||||
|
||||
async def _refresh():
|
||||
refresh_calls.append("refresh")
|
||||
|
||||
monkeypatch.setattr("app.gateway.routers.skills.refresh_skills_system_prompt_cache_async", _refresh)
|
||||
|
||||
app = FastAPI()
|
||||
app.include_router(skills_router.router)
|
||||
|
||||
with TestClient(app) as client:
|
||||
response = client.get("/api/skills/custom")
|
||||
assert response.status_code == 200
|
||||
assert response.json()["skills"][0]["name"] == "demo-skill"
|
||||
|
||||
get_response = client.get("/api/skills/custom/demo-skill")
|
||||
assert get_response.status_code == 200
|
||||
assert "# demo-skill" in get_response.json()["content"]
|
||||
|
||||
update_response = client.put(
|
||||
"/api/skills/custom/demo-skill",
|
||||
json={"content": _skill_content("demo-skill", "Edited skill")},
|
||||
)
|
||||
assert update_response.status_code == 200
|
||||
assert update_response.json()["description"] == "Edited skill"
|
||||
|
||||
history_response = client.get("/api/skills/custom/demo-skill/history")
|
||||
assert history_response.status_code == 200
|
||||
assert history_response.json()["history"][-1]["action"] == "human_edit"
|
||||
|
||||
rollback_response = client.post("/api/skills/custom/demo-skill/rollback", json={"history_index": -1})
|
||||
assert rollback_response.status_code == 200
|
||||
assert rollback_response.json()["description"] == "Demo skill"
|
||||
assert refresh_calls == ["refresh", "refresh"]
|
||||
|
||||
|
||||
def test_custom_skill_rollback_blocked_by_scanner(monkeypatch, tmp_path):
|
||||
skills_root = tmp_path / "skills"
|
||||
custom_dir = skills_root / "custom" / "demo-skill"
|
||||
custom_dir.mkdir(parents=True, exist_ok=True)
|
||||
original_content = _skill_content("demo-skill")
|
||||
edited_content = _skill_content("demo-skill", "Edited skill")
|
||||
(custom_dir / "SKILL.md").write_text(edited_content, encoding="utf-8")
|
||||
config = SimpleNamespace(
|
||||
skills=SimpleNamespace(get_skills_path=lambda: skills_root, container_path="/mnt/skills"),
|
||||
skill_evolution=SimpleNamespace(enabled=True, moderation_model_name=None),
|
||||
)
|
||||
monkeypatch.setattr("deerflow.config.get_app_config", lambda: config)
|
||||
monkeypatch.setattr("deerflow.skills.manager.get_app_config", lambda: config)
|
||||
get_skill_history_file("demo-skill").write_text(
|
||||
'{"action":"human_edit","prev_content":' + json.dumps(original_content) + ',"new_content":' + json.dumps(edited_content) + "}\n",
|
||||
encoding="utf-8",
|
||||
)
|
||||
|
||||
async def _refresh():
|
||||
return None
|
||||
|
||||
monkeypatch.setattr("app.gateway.routers.skills.refresh_skills_system_prompt_cache_async", _refresh)
|
||||
|
||||
async def _scan(*args, **kwargs):
|
||||
from deerflow.skills.security_scanner import ScanResult
|
||||
|
||||
return ScanResult(decision="block", reason="unsafe rollback")
|
||||
|
||||
monkeypatch.setattr("app.gateway.routers.skills.scan_skill_content", _scan)
|
||||
|
||||
app = FastAPI()
|
||||
app.include_router(skills_router.router)
|
||||
|
||||
with TestClient(app) as client:
|
||||
rollback_response = client.post("/api/skills/custom/demo-skill/rollback", json={"history_index": -1})
|
||||
assert rollback_response.status_code == 400
|
||||
assert "unsafe rollback" in rollback_response.json()["detail"]
|
||||
|
||||
history_response = client.get("/api/skills/custom/demo-skill/history")
|
||||
assert history_response.status_code == 200
|
||||
assert history_response.json()["history"][-1]["scanner"]["decision"] == "block"
|
||||
|
||||
|
||||
def test_custom_skill_delete_preserves_history_and_allows_restore(monkeypatch, tmp_path):
|
||||
skills_root = tmp_path / "skills"
|
||||
custom_dir = skills_root / "custom" / "demo-skill"
|
||||
custom_dir.mkdir(parents=True, exist_ok=True)
|
||||
original_content = _skill_content("demo-skill")
|
||||
(custom_dir / "SKILL.md").write_text(original_content, encoding="utf-8")
|
||||
config = SimpleNamespace(
|
||||
skills=SimpleNamespace(get_skills_path=lambda: skills_root, container_path="/mnt/skills"),
|
||||
skill_evolution=SimpleNamespace(enabled=True, moderation_model_name=None),
|
||||
)
|
||||
monkeypatch.setattr("deerflow.config.get_app_config", lambda: config)
|
||||
monkeypatch.setattr("deerflow.skills.manager.get_app_config", lambda: config)
|
||||
monkeypatch.setattr("app.gateway.routers.skills.scan_skill_content", lambda *args, **kwargs: _async_scan("allow", "ok"))
|
||||
refresh_calls = []
|
||||
|
||||
async def _refresh():
|
||||
refresh_calls.append("refresh")
|
||||
|
||||
monkeypatch.setattr("app.gateway.routers.skills.refresh_skills_system_prompt_cache_async", _refresh)
|
||||
|
||||
app = FastAPI()
|
||||
app.include_router(skills_router.router)
|
||||
|
||||
with TestClient(app) as client:
|
||||
delete_response = client.delete("/api/skills/custom/demo-skill")
|
||||
assert delete_response.status_code == 200
|
||||
assert not (custom_dir / "SKILL.md").exists()
|
||||
|
||||
history_response = client.get("/api/skills/custom/demo-skill/history")
|
||||
assert history_response.status_code == 200
|
||||
assert history_response.json()["history"][-1]["action"] == "human_delete"
|
||||
|
||||
rollback_response = client.post("/api/skills/custom/demo-skill/rollback", json={"history_index": -1})
|
||||
assert rollback_response.status_code == 200
|
||||
assert rollback_response.json()["description"] == "Demo skill"
|
||||
assert (custom_dir / "SKILL.md").read_text(encoding="utf-8") == original_content
|
||||
assert refresh_calls == ["refresh", "refresh"]
|
||||
|
||||
|
||||
def test_update_skill_refreshes_prompt_cache_before_return(monkeypatch, tmp_path):
|
||||
config_path = tmp_path / "extensions_config.json"
|
||||
enabled_state = {"value": True}
|
||||
refresh_calls = []
|
||||
|
||||
def _load_skills(*, enabled_only: bool):
|
||||
skill = _make_skill("demo-skill", enabled=enabled_state["value"])
|
||||
if enabled_only and not skill.enabled:
|
||||
return []
|
||||
return [skill]
|
||||
|
||||
async def _refresh():
|
||||
refresh_calls.append("refresh")
|
||||
enabled_state["value"] = False
|
||||
|
||||
monkeypatch.setattr("app.gateway.routers.skills.load_skills", _load_skills)
|
||||
monkeypatch.setattr("app.gateway.routers.skills.get_extensions_config", lambda: SimpleNamespace(mcp_servers={}, skills={}))
|
||||
monkeypatch.setattr("app.gateway.routers.skills.reload_extensions_config", lambda: None)
|
||||
monkeypatch.setattr(skills_router.ExtensionsConfig, "resolve_config_path", staticmethod(lambda: config_path))
|
||||
monkeypatch.setattr("app.gateway.routers.skills.refresh_skills_system_prompt_cache_async", _refresh)
|
||||
|
||||
app = FastAPI()
|
||||
app.include_router(skills_router.router)
|
||||
|
||||
with TestClient(app) as client:
|
||||
response = client.put("/api/skills/demo-skill", json={"enabled": False})
|
||||
|
||||
assert response.status_code == 200
|
||||
assert response.json()["enabled"] is False
|
||||
assert refresh_calls == ["refresh"]
|
||||
assert json.loads(config_path.read_text(encoding="utf-8")) == {"mcpServers": {}, "skills": {"demo-skill": {"enabled": False}}}
|
||||
Reference in New Issue
Block a user