open-swe/agent/utils/model.py
Johannes du Plessis 88856a04fa
feat: open-swe dashboard for per-user profile config (#1302)
* feat: dashboard backend — GitHub OAuth, profile CRUD, admin endpoints

Adds agent/dashboard/ FastAPI router mounted at /dashboard/api covering:
- GitHub App OAuth login → JWT cookie session (cross-domain ready)
- profile CRUD against LangGraph Store with model+effort validation
- admin gate via CONFIGURED_ADMINS
- /repos via /user/installations using the user's encrypted OAuth token

CORS allowlist on webapp.py is opt-in via DASHBOARD_ALLOWED_ORIGINS so the
Vercel-hosted frontend can call the LangSmith deployment with credentials.

* feat: apply dashboard profile model/effort overrides in get_agent

Look up the triggering user's GitHub login from config (direct field or
GITHUB_USER_EMAIL_MAP reverse lookup), read their profile from the Store,
and apply default_model + reasoning_effort to make_model when both are
valid. Effort 'max' is captured on the profile but not yet wired through —
the OpenAI Reasoning Literal doesn't accept it.

* feat: ui/ TanStack Start dashboard for profile config

Scaffolded with the shadcn b7CScJIjA preset (TanStack Start template,
base-ui primitives, Tailwind v4). Three routes:

- /login   — Sign in with GitHub (links to /dashboard/api/auth/login)
- /profile — Edit default model, reasoning effort, default repo
- /admin   — Admin-only: list users and edit other profiles

API client (src/lib/api.ts) uses credentials: include so the osw_session
cookie set by the OAuth callback rides cross-origin. VITE_DASHBOARD_API_BASE_URL
points at the LangSmith deployment.

Effort options re-render when the model changes; 'max' on Opus 4.7 is
captured on the profile but ignored downstream until anthropic reasoning
is wired through make_model.

* feat: searchable Combobox for default repo picker

Replaces the Select with a base-ui Combobox so users can filter by typing,
the popup is wider than the trigger so full owner/repo names are readable,
and the list caps at max-h-80 to stay on screen.

* fix: address review comments + wire default_repo and Anthropic thinking

Security/correctness fixes from PR review:

* Open redirect: validate `redirect_to` in `/auth/login` against
  `DASHBOARD_BASE_URL` + `DASHBOARD_ALLOWED_ORIGINS` before signing it
  into the state JWT. Anything off-allowlist falls back to the dashboard
  base URL. (PR #1302 r3250054386)

* Login CSRF: bind the OAuth `state` to the requesting browser. At
  `/auth/login` we generate a fresh nonce, set it as a short-lived
  HttpOnly SameSite=Lax cookie scoped to `/dashboard/api/auth`, and
  embed `hash_state_nonce(nonce)` in the state JWT. At `/auth/callback`
  we require the cookie nonce to hash-match the state JWT's nonce_hash
  (constant-time compare). (PR #1302 r3250054395)

* RMW race in profile vs token writes: split storage into two
  namespaces — `["profiles"]` for user-editable settings and
  `["oauth_tokens"]` for the encrypted GitHub token. Each upsert now
  only writes its own namespace so an in-flight profile save can no
  longer clobber a fresh token from a concurrent re-login (and vice
  versa). (PR #1302 r3250054393)

* /repos pagination: follow `Link: rel="next"` for both
  `/user/installations` and per-installation `/repositories` with
  per_page=100, capped at 1000 items. (PR #1302 r3250054401)

Feature wires:

* default_repo: applied as a fallback in `get_slack_repo_config` (after
  explicit-repo / thread metadata, before the env defaults) and in the
  Linear webhook (after comment-body extraction, before team mapping).
  Both paths resolve the triggering user's GitHub login via
  GITHUB_USER_EMAIL_MAP and read the profile's default_repo.

* Anthropic "thinking" effort: `make_model` now accepts a `thinking`
  kwarg; `get_agent` maps profile effort {low,medium,high,xhigh,max}
  to budget_tokens {1k,4k,12k,32k,60k} when the chosen model is
  anthropic. OpenAI path still ignores "max" since the Literal doesn't
  accept it.
2026-05-15 11:23:53 -07:00

55 lines
1.8 KiB
Python

from typing import Literal, TypedDict, Unpack
from langchain.chat_models import init_chat_model
OPENAI_RESPONSES_WS_BASE_URL = "wss://api.openai.com/v1"
# Anthropic SDK default is 2; a 529 burst can outlive that. Bump to give the
# primary provider a fair chance before the fallback middleware kicks in.
DEFAULT_MAX_RETRIES = 6
OpenAIReasoningEffort = Literal["none", "low", "medium", "high", "xhigh"]
AnthropicThinkingType = Literal["enabled", "disabled"]
class OpenAIReasoning(TypedDict, total=False):
effort: OpenAIReasoningEffort
class AnthropicThinking(TypedDict, total=False):
type: AnthropicThinkingType
budget_tokens: int
class ModelKwargs(TypedDict, total=False):
max_tokens: int | None
reasoning: OpenAIReasoning | None
thinking: AnthropicThinking | None
temperature: float | None
max_retries: int | None
def make_model(model_id: str, **kwargs: Unpack[ModelKwargs]):
model_kwargs: dict[str, object] = kwargs.copy()
model_kwargs.setdefault("max_retries", DEFAULT_MAX_RETRIES)
if model_id.startswith("openai:"):
model_kwargs["base_url"] = OPENAI_RESPONSES_WS_BASE_URL
model_kwargs["use_responses_api"] = True
return init_chat_model(model=model_id, **model_kwargs)
def fallback_model_id_for(primary_model_id: str) -> str | None:
"""Return the cross-provider fallback model id for a given primary, if any.
Anthropic primaries fall back to OpenAI and vice versa. Returns ``None``
when the provider has no configured cross-provider fallback (e.g. local
or self-hosted providers we don't want to silently route off-host).
"""
if primary_model_id.startswith("anthropic:"):
return "openai:gpt-5.5"
if primary_model_id.startswith("openai:"):
return "anthropic:claude-opus-4-5"
return None