create-workframe 0.1.4 → 0.1.6

package/workframe-api/test_billing_provider.py

@@ -0,0 +1,24 @@
+"""ponytail self-check: billing provider source of truth = proxy URL, never model-id prefix.
+
+Run: python services/workframe-api/test_billing_provider.py
+No framework, no fixtures. Fails fast if the inference regression returns.
+"""
+import sys
+from pathlib import Path
+
+sys.path.insert(0, str(Path(__file__).resolve().parent))
+
+import server  # noqa: E402  (module import is side-effect-free; main() is guarded)
+
+# The bug case: profile model is google/gemini-2.5-flash but proxy routes via openrouter.
+# The model-id prefix MUST NOT leak as the billing provider.
+assert server._billing_provider_from_block("/internal/llm/openrouter/v1", "custom", "") == "openrouter", \
+    "proxy URL segment must win over model-id prefix"
+# Explicit hint honored.
+assert server._billing_provider_from_block("", "", "anthropic") == "anthropic", "hint honored"
+# Empty block defaults to openrouter (no inference).
+assert server._billing_provider_from_block("", "", "") == "openrouter", "empty-block default"
+# cfg_provider honored only when no proxy URL is set.
+assert server._billing_provider_from_block("", "openai", "") == "openai", "cfg_provider honored without proxy"
+
+print("billing provider self-check ok")

package/workframe-api/time-bind-chat.py

@@ -1,99 +1,99 @@
-#!/usr/bin/env python3
-"""Time bind + credential lease on running workframe-api (docker exec)."""
-from __future__ import annotations
-
-import sqlite3
-import sys
-import time
-
-import server
-
-
-def main() -> int:
-    conn = sqlite3.connect(str(server.DATA_DIR / "workframe.db"))
-    conn.row_factory = sqlite3.Row
-    room = conn.execute(
-        """
-        SELECT r.id, r.workspace_id, r.agent_profile_id, ap.slug AS agent_slug
-        FROM rooms r
-        JOIN agent_profiles ap ON ap.id = r.agent_profile_id
-        JOIN room_memberships rm ON rm.room_id = r.id AND rm.status = 'active'
-        WHERE r.room_type = 'direct' AND r.deleted_at IS NULL
-          AND ap.slug = 'workframe-agent'
-        ORDER BY r.updated_at DESC
-        LIMIT 1
-        """,
-    ).fetchone()
-    if not room:
-        print("no workframe-agent room", file=sys.stderr)
-        return 1
-    user = conn.execute(
-        """
-        SELECT u.id FROM users u
-        JOIN room_memberships rm ON rm.user_id = u.id AND rm.room_id = ?
-        WHERE u.status = 'active'
-        LIMIT 1
-        """,
-        (str(room["id"]),),
-    ).fetchone()
-    conn.close()
-    if not room or not user:
-        print("no room/user", file=sys.stderr)
-        return 1
-
-    user_id = str(user["id"])
-    room_id = str(room["id"])
-    workspace_id = str(room["workspace_id"])
-    payload = {
-        "room_id": room_id,
-        "workspace_id": workspace_id,
-        "source_id": "timing",
-        "client_id": "timing",
-        "binding_version": 2,
-    }
-
-    t_rt = time.perf_counter()
-    prof = server._resolve_chat_hermes_profile("workframe-agent", user_id, room_id, workspace_id)
-    rt_ms = (time.perf_counter() - t_rt) * 1000
-    print(f"resolve_runtime_ms={rt_ms:.0f} profile={prof}")
-
-    provider = server._llm_billing_provider(prof)
-    t_api = time.perf_counter()
-    server.ensure_profile_api(prof, user_id, workspace_id)
-    api_ms = (time.perf_counter() - t_api) * 1000
-    t_llm = time.perf_counter()
-    server._overlay_chat_llm_env(prof, user_id, workspace_id, provider)
-    llm_ms = (time.perf_counter() - t_llm) * 1000
-    print(f"ensure_profile_api_ms={api_ms:.0f} overlay_llm_ms={llm_ms:.0f}")
-
-    t0 = time.perf_counter()
-    session = server.profile_chat_session("workframe-agent", payload, user_id=user_id)
-    session_ms = (time.perf_counter() - t0) * 1000
-    sid = str(session.get("session_id") or "")
-    print(f"session_ms={session_ms:.0f} session={sid[:40]}")
-
-    t_hist = time.perf_counter()
-    history = server.chat_messages(prof, sid)
-    hist_ms = (time.perf_counter() - t_hist) * 1000
-    print(f"history_ms={hist_ms:.0f} messages={len(history.get('messages') or [])}")
-
-    healthy = server._profile_api_healthy(prof)
-    print(f"profile_api_healthy={healthy} port={server._profile_api_port(prof)}")
-
-    run_a = "timing-run-a"
-    run_b = "timing-run-b"
-    t1 = time.perf_counter()
-    server._apply_turn_credential_lease(prof, user_id, workspace_id, provider, run_a)
-    lease1_ms = (time.perf_counter() - t1) * 1000
-    t2 = time.perf_counter()
-    server._apply_turn_credential_lease(prof, user_id, workspace_id, provider, run_b)
-    lease2_ms = (time.perf_counter() - t2) * 1000
-    print(f"lease_cold_ms={lease1_ms:.0f} lease_reuse_ms={lease2_ms:.0f} provider={provider}")
-
-    _, model = server._read_model_from_config(prof)
-    print(f"model={model}")
-    return 0
-
-
-if __name__ == "__main__":
-    raise SystemExit(main())
+#!/usr/bin/env python3
+"""Time bind + credential lease on running workframe-api (docker exec)."""
+from __future__ import annotations
+
+import sqlite3
+import sys
+import time
+
+import server
+
+
+def main() -> int:
+    conn = sqlite3.connect(str(server.DATA_DIR / "workframe.db"))
+    conn.row_factory = sqlite3.Row
+    room = conn.execute(
+        """
+        SELECT r.id, r.workspace_id, r.agent_profile_id, ap.slug AS agent_slug
+        FROM rooms r
+        JOIN agent_profiles ap ON ap.id = r.agent_profile_id
+        JOIN room_memberships rm ON rm.room_id = r.id AND rm.status = 'active'
+        WHERE r.room_type = 'direct' AND r.deleted_at IS NULL
+          AND ap.slug = 'workframe-agent'
+        ORDER BY r.updated_at DESC
+        LIMIT 1
+        """,
+    ).fetchone()
+    if not room:
+        print("no workframe-agent room", file=sys.stderr)
+        return 1
+    user = conn.execute(
+        """
+        SELECT u.id FROM users u
+        JOIN room_memberships rm ON rm.user_id = u.id AND rm.room_id = ?
+        WHERE u.status = 'active'
+        LIMIT 1
+        """,
+        (str(room["id"]),),
+    ).fetchone()
+    conn.close()
+    if not room or not user:
+        print("no room/user", file=sys.stderr)
+        return 1
+
+    user_id = str(user["id"])
+    room_id = str(room["id"])
+    workspace_id = str(room["workspace_id"])
+    payload = {
+        "room_id": room_id,
+        "workspace_id": workspace_id,
+        "source_id": "timing",
+        "client_id": "timing",
+        "binding_version": 2,
+    }
+
+    t_rt = time.perf_counter()
+    prof = server._resolve_chat_hermes_profile("workframe-agent", user_id, room_id, workspace_id)
+    rt_ms = (time.perf_counter() - t_rt) * 1000
+    print(f"resolve_runtime_ms={rt_ms:.0f} profile={prof}")
+
+    provider = server._llm_billing_provider(prof)
+    t_api = time.perf_counter()
+    server.ensure_profile_api(prof, user_id, workspace_id)
+    api_ms = (time.perf_counter() - t_api) * 1000
+    t_llm = time.perf_counter()
+    server._overlay_chat_llm_env(prof, user_id, workspace_id, provider)
+    llm_ms = (time.perf_counter() - t_llm) * 1000
+    print(f"ensure_profile_api_ms={api_ms:.0f} overlay_llm_ms={llm_ms:.0f}")
+
+    t0 = time.perf_counter()
+    session = server.profile_chat_session("workframe-agent", payload, user_id=user_id)
+    session_ms = (time.perf_counter() - t0) * 1000
+    sid = str(session.get("session_id") or "")
+    print(f"session_ms={session_ms:.0f} session={sid[:40]}")
+
+    t_hist = time.perf_counter()
+    history = server.chat_messages(prof, sid)
+    hist_ms = (time.perf_counter() - t_hist) * 1000
+    print(f"history_ms={hist_ms:.0f} messages={len(history.get('messages') or [])}")
+
+    healthy = server._profile_api_healthy(prof)
+    print(f"profile_api_healthy={healthy} port={server._profile_api_port(prof)}")
+
+    run_a = "timing-run-a"
+    run_b = "timing-run-b"
+    t1 = time.perf_counter()
+    server._apply_turn_credential_lease(prof, user_id, workspace_id, provider, run_a)
+    lease1_ms = (time.perf_counter() - t1) * 1000
+    t2 = time.perf_counter()
+    server._apply_turn_credential_lease(prof, user_id, workspace_id, provider, run_b)
+    lease2_ms = (time.perf_counter() - t2) * 1000
+    print(f"lease_cold_ms={lease1_ms:.0f} lease_reuse_ms={lease2_ms:.0f} provider={provider}")
+
+    _, model = server._read_model_from_config(prof)
+    print(f"model={model}")
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())