Phase 5: multi-host agents (0.5.0)

- stackpilot-agent: slim token-guarded FastAPI (reuses compose_service) exposing
  stack CRUD/lifecycle/logs + system info; same image, different CMD. agent/
  Dockerfile + compose + .env.example.
- Central proxy: Agent model, agent_service (httpx ping/proxy + live status:
  online/offline/unauthorized + hostname/last_seen), routers/agents.py
  (CRUD + ping + proxied stacks/lifecycle/logs/system).
- Frontend: Settings → Remote hosts (add/check/remove, connectivity dot); Stacks
  grouped by host; remote stack detail with lifecycle, live logs, compose/.env edit.

Verified end-to-end: agent+main on a shared network — register (good/bad token),
list/create/start/logs/delete remote stacks, offline detection (502).

Remote backup destinations (SFTP/S3) deferred.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
menzelj
2026-06-07 21:23:17 +00:00
co-authored by Claude Opus 4.8
parent 8d19b09abd
commit 59037f4287
21 changed files with 1380 additions and 39 deletions
+286
View File
@@ -0,0 +1,286 @@
"""Remote host (agent) management + proxied stack/system operations."""
from __future__ import annotations
from fastapi import APIRouter, Depends, HTTPException, Query, Request
from sqlmodel import Session, select
from auth import get_current_user, require_admin
from database import get_session
from models.agent import Agent, AgentCreate, AgentRead, AgentUpdate
from models.stack import StackCreate, StackUpdate
from models.user import User
from services import agent_service, audit_service
from services.agent_service import AgentError
router = APIRouter(prefix="/api/agents", tags=["agents"])
_ACTIONS = {"start", "stop", "restart", "pull", "update", "down"}
def _ip(request: Request) -> str:
return request.client.host if request.client else "unknown"
def _to_read(a: Agent) -> AgentRead:
return AgentRead(
id=a.id,
name=a.name,
url=a.url,
status=a.status,
hostname=a.hostname,
last_seen=a.last_seen,
created_at=a.created_at,
token_set=bool(a.token),
)
def _get_or_404(session: Session, agent_id: int) -> Agent:
agent = session.get(Agent, agent_id)
if not agent:
raise HTTPException(status_code=404, detail=f"Agent {agent_id} not found")
return agent
def _raise(exc: AgentError):
raise HTTPException(
status_code=exc.status if exc.status >= 400 else 502,
detail={"error": exc.error, "detail": exc.detail},
)
async def _proxy(session: Session, agent: Agent, method: str, path: str, **kw):
try:
return await agent_service.call(session, agent, method, path, **kw)
except AgentError as exc:
_raise(exc)
# --------------------------------------------------------------------------- #
# CRUD
# --------------------------------------------------------------------------- #
@router.get("", response_model=list[AgentRead])
async def list_agents(
refresh: bool = Query(True),
session: Session = Depends(get_session),
_user: User = Depends(get_current_user),
) -> list[AgentRead]:
agents = session.exec(select(Agent).order_by(Agent.id)).all()
if refresh:
for agent in agents:
await agent_service.ping(session, agent)
return [_to_read(a) for a in agents]
@router.post("", response_model=AgentRead, status_code=201)
async def create_agent(
body: AgentCreate,
request: Request,
session: Session = Depends(get_session),
user: User = Depends(require_admin),
) -> AgentRead:
agent = Agent(name=body.name, url=body.url.rstrip("/"), token=body.token)
session.add(agent)
session.commit()
session.refresh(agent)
# Validate connectivity immediately (best-effort; agent is saved regardless).
await agent_service.ping(session, agent)
audit_service.record(
session, user=user.username, action="agent.create", target=agent.name,
detail=agent.url, ip=_ip(request),
)
return _to_read(agent)
@router.put("/{agent_id}", response_model=AgentRead)
async def update_agent(
agent_id: int,
body: AgentUpdate,
request: Request,
session: Session = Depends(get_session),
user: User = Depends(require_admin),
) -> AgentRead:
agent = _get_or_404(session, agent_id)
if body.name is not None:
agent.name = body.name
if body.url is not None:
agent.url = body.url.rstrip("/")
if body.token:
agent.token = body.token
agent.status = "unknown"
session.add(agent)
session.commit()
session.refresh(agent)
await agent_service.ping(session, agent)
audit_service.record(
session, user=user.username, action="agent.update", target=agent.name,
ip=_ip(request),
)
return _to_read(agent)
@router.delete("/{agent_id}")
def delete_agent(
agent_id: int,
request: Request,
session: Session = Depends(get_session),
user: User = Depends(require_admin),
) -> dict:
agent = _get_or_404(session, agent_id)
name = agent.name
session.delete(agent)
session.commit()
audit_service.record(
session, user=user.username, action="agent.delete", target=name, ip=_ip(request),
)
return {"ok": True}
@router.post("/{agent_id}/ping")
async def ping_agent(
agent_id: int,
session: Session = Depends(get_session),
_user: User = Depends(get_current_user),
) -> dict:
agent = _get_or_404(session, agent_id)
return await agent_service.ping(session, agent)
# --------------------------------------------------------------------------- #
# Proxied stack + system operations
# --------------------------------------------------------------------------- #
@router.get("/{agent_id}/system")
async def agent_system(
agent_id: int,
session: Session = Depends(get_session),
_user: User = Depends(get_current_user),
):
agent = _get_or_404(session, agent_id)
return await _proxy(session, agent, "GET", "/agent/system")
@router.get("/{agent_id}/stacks")
async def agent_stacks(
agent_id: int,
session: Session = Depends(get_session),
_user: User = Depends(get_current_user),
) -> list[dict]:
agent = _get_or_404(session, agent_id)
stacks = await _proxy(session, agent, "GET", "/agent/stacks") or []
for s in stacks:
s["agent_id"] = agent.id
s["agent_name"] = agent.name
return stacks
@router.get("/{agent_id}/stacks/{stack_id}")
async def agent_stack_detail(
agent_id: int,
stack_id: str,
session: Session = Depends(get_session),
_user: User = Depends(get_current_user),
) -> dict:
agent = _get_or_404(session, agent_id)
data = await _proxy(session, agent, "GET", f"/agent/stacks/{stack_id}")
data["agent_id"] = agent.id
data["agent_name"] = agent.name
return data
@router.get("/{agent_id}/stacks/{stack_id}/logs")
async def agent_stack_logs(
agent_id: int,
stack_id: str,
tail: int = Query(200, le=2000),
session: Session = Depends(get_session),
_user: User = Depends(get_current_user),
) -> dict:
agent = _get_or_404(session, agent_id)
return await _proxy(
session, agent, "GET", f"/agent/stacks/{stack_id}/logs", params={"tail": tail}
)
@router.post("/{agent_id}/stacks", status_code=201)
async def agent_create_stack(
agent_id: int,
body: StackCreate,
request: Request,
session: Session = Depends(get_session),
user: User = Depends(require_admin),
) -> dict:
agent = _get_or_404(session, agent_id)
result = await _proxy(
session, agent, "POST", "/agent/stacks",
json={"name": body.name, "yaml": body.yaml, "env": body.env},
)
audit_service.record(
session, user=user.username, action="agent.stack.create",
target=f"{agent.name}/{result.get('id')}", ip=_ip(request),
)
return result
@router.put("/{agent_id}/stacks/{stack_id}")
async def agent_update_stack(
agent_id: int,
stack_id: str,
body: StackUpdate,
request: Request,
session: Session = Depends(get_session),
user: User = Depends(require_admin),
) -> dict:
agent = _get_or_404(session, agent_id)
result = await _proxy(
session, agent, "PUT", f"/agent/stacks/{stack_id}",
json={"yaml": body.yaml, "env": body.env},
)
audit_service.record(
session, user=user.username, action="agent.stack.update",
target=f"{agent.name}/{stack_id}", ip=_ip(request),
)
return result
@router.delete("/{agent_id}/stacks/{stack_id}")
async def agent_delete_stack(
agent_id: int,
stack_id: str,
request: Request,
delete_files: bool = Query(True),
session: Session = Depends(get_session),
user: User = Depends(require_admin),
) -> dict:
agent = _get_or_404(session, agent_id)
result = await _proxy(
session, agent, "DELETE", f"/agent/stacks/{stack_id}",
params={"delete_files": delete_files},
)
audit_service.record(
session, user=user.username, action="agent.stack.delete",
target=f"{agent.name}/{stack_id}", ip=_ip(request),
)
return result
@router.post("/{agent_id}/stacks/{stack_id}/{action}")
async def agent_lifecycle(
agent_id: int,
stack_id: str,
action: str,
request: Request,
session: Session = Depends(get_session),
user: User = Depends(require_admin),
) -> dict:
if action not in _ACTIONS:
raise HTTPException(status_code=400, detail=f"Unknown action '{action}'")
agent = _get_or_404(session, agent_id)
result = await _proxy(session, agent, "POST", f"/agent/stacks/{stack_id}/{action}")
audit_service.record(
session, user=user.username, action=f"agent.stack.{action}",
target=f"{agent.name}/{stack_id}", ip=_ip(request),
)
return result