"""Read-only gateway introspection commands: /status, /context, /usage, /agents, /insights, /topup. Bound onto ``GatewayRunner`` through ``GatewaySlashCommandsMixin``.""" from __future__ import annotations import logging import asyncio import hashlib import os import re import time from typing import Any from agent.account_usage import fetch_account_usage, render_account_usage_lines from agent.i18n import t from gateway.config import Platform from gateway.platforms.event import MessageEvent from gateway.session_transcript import TranscriptReadError from hermes_cli.status_report import build_status_fields # Log-record parity with gateway/run.py and the origin module. logger = logging.getLogger("gateway.run") _LIST_CAP = 12 # /agents shows at most this many rows per section def _clean_str(value: Any) -> str: """Strip and return a non-empty string value, or empty string.""" return value.strip() if isinstance(value, str) and value.strip() else "" def _int_value(value: Any) -> int: """Safely coerce to int.""" try: return int(value) except (TypeError, ValueError): return 0 def _n(obj, attr: str): return getattr(obj, attr, 0) or 0 def _fmt(n) -> str: return f"{n:,}" def _pct(used, total) -> float: # clamped occupancy percentage; 0 for an unknown window return min(100, used / total * 100) if total else 0 def _clip(text: str, limit: int) -> str: return text[: limit - 3] + "..." if len(text) > limit else text def _transcript_estimate(history) -> tuple[int, int]: """``(approx_tokens, message_count)`` over the user/assistant messages of a transcript.""" from agent.model_metadata import estimate_messages_tokens_rough msgs = [m for m in history if m.get("role") in {"user", "assistant"} and m.get("content")] return estimate_messages_tokens_rough(msgs), len(msgs) async def _quiet(call, default=None): """Await ``call()`` fail-open: any exception (sync or in the awaitable) yields *default*.""" try: return await call() except Exception: return default # English source for the transcript-read-failure reply; callers render ``history_unreadable()`` # (``gateway.shared.history_unreadable``) so the active language applies. HISTORY_UNREADABLE = ("⚠️ I can't read this conversation's history right now (your earlier messages " "exist but cannot be loaded). Run `hermes doctor --fix` on the host, or use /new " "to start fresh.") def history_unreadable() -> str: return t("gateway.shared.history_unreadable") def _configured_provider() -> str: """``model.provider`` from the gateway config ("" when unset).""" from gateway.run import _load_gateway_config user_config = _load_gateway_config() model_cfg = user_config.get("model", {}) if isinstance(user_config, dict) else {} return _clean_str(model_cfg.get("provider")) if isinstance(model_cfg, dict) else "" def _quiet_sync(call, default=None): """Sync twin of ``_quiet``.""" try: return call() except Exception: return default def _status_model_route( status_agent, active_override: dict, persisted_route: dict, session_row: dict, session_entry ): """``(model, provider, context_used, context_total, route)`` for /status. Order: live/cached agent route -> active session override -> persisted recent route -> SessionDB row -> gateway config (only loaded when something is still missing). ``route`` carries the ``provider`` / ``base_url`` / ``api_key`` of the winning source only, so a later context-window lookup queries the endpoint that serves the displayed model (never a losing route's endpoint); a winner without a ``base_url`` leaves the lookup on the default runtime route. """ from gateway.run import _AGENT_PENDING_SENTINEL, _load_gateway_config, _resolve_gateway_model context_used = context_total = 0 routes: list[tuple[str, str, dict]] = [] if status_agent is not None and status_agent is not _AGENT_PENDING_SENTINEL: routes.append((_clean_str(getattr(status_agent, "model", "")), _clean_str(getattr(status_agent, "provider", "")), {"base_url": _clean_str(getattr(status_agent, "base_url", "")), "api_key": _clean_str(getattr(status_agent, "api_key", ""))})) ctx = getattr(status_agent, "context_compressor", None) if ctx is not None: context_used = max(0, _int_value(getattr(ctx, "last_prompt_tokens", 0))) context_total = _int_value(getattr(ctx, "context_length", 0)) routes.append((_clean_str(active_override.get("model")), _clean_str(active_override.get("provider")), {"base_url": _clean_str(active_override.get("base_url")), "api_key": _clean_str(active_override.get("api_key"))})) routes.append((_clean_str(persisted_route.get("model")), _clean_str(persisted_route.get("billing_provider")), {})) row_route = (_clean_str(session_row.get("model")), _clean_str(session_row.get("billing_provider")), {}) # First fully-resolved (model AND provider) route wins; the SessionDB row is used even if partial. model_name, provider_name, route = next((r for r in routes if r[0] and r[1]), row_route) context_used = context_used or _int_value(getattr(session_entry, "last_prompt_tokens", 0)) user_config: dict[str, Any] = {} if not model_name or not provider_name: user_config = _quiet_sync(_load_gateway_config, {}) model_cfg = user_config.get("model", {}) if isinstance(user_config, dict) else {} model_cfg = model_cfg if isinstance(model_cfg, dict) else {} model_name = model_name or _resolve_gateway_model(user_config) provider_name = provider_name or _clean_str(model_cfg.get("provider")) # No raw ``model.context_length`` pin here: the resolver applies it only while the displayed # route still matches the configured one (a session /model switch must not inherit it). return model_name, provider_name, context_used, context_total, {"provider": provider_name, **route} def _context_compressor_lines(agent, ctx, used: int) -> list[str]: """/context full view: auto-compression threshold/headroom, compression count + last savings, and cumulative throughput (labelled as throughput, NOT context size).""" lines: list[str] = [] from agent.context_breakdown import context_display_source mark = "~" if context_display_source(ctx) != "provider_usage" else "" threshold = _n(ctx, "threshold_tokens") threshold_pct = f"{_n(ctx, 'threshold_percent') * 100:.0f}" if threshold > 0: if used >= threshold: lines.append(t("gateway.context.over_threshold", threshold=_fmt(threshold), threshold_pct=threshold_pct)) else: lines.append(t("gateway.context.threshold", threshold=_fmt(threshold), threshold_pct=threshold_pct, to_go=mark + _fmt(threshold - used))) compressions = _n(ctx, "compression_count") lines.append(t("gateway.context.compressions", count=compressions)) savings = getattr(ctx, "_last_compression_savings_pct", None) if compressions else None if savings is not None: lines.append(t("gateway.context.last_savings", savings=f"{savings:.0f}")) lines += [ "", t("gateway.context.totals_header", calls=_n(agent, "session_api_calls")), t("gateway.context.totals_line", input=_fmt(_n(agent, "session_input_tokens")), output=_fmt(_n(agent, "session_output_tokens")), reasoning=_fmt(_n(agent, "session_reasoning_tokens"))), t("gateway.context.total_billed", total=_fmt(_n(agent, "session_total_tokens"))), t("gateway.context.throughput_note"), ] return lines def _agents_delegation_lines(d: dict) -> list[str]: """/agents rows for one background delegation. Live per-child activity comes from the registry's progress sampler: api calls, current tool, seconds since last activity.""" goal = _clip(" ".join(str(d.get("goal") or "").split()), 70) status = d.get("status", "?") row = f"- `{d.get('delegation_id', '?')}` · {status}" quiet = d.get("stalled_after_quiet_seconds") if status == "stalling" and quiet is not None: row += t("gateway.agents.no_progress", seconds=f"{quiet:.0f}") elif status != "stalling" and d.get("seconds_since_progress", 0) >= 60: row += t("gateway.agents.quiet_for", seconds=f"{d['seconds_since_progress']:.0f}") if goal: row += f" · {goal}" lines = [row] for i, child in enumerate(d.get("children_activity") or []): if not isinstance(child, dict): continue tool = child.get("current_tool") doing = f"`{tool}`" if tool else t("gateway.agents.between_turns") part = t("gateway.agents.child_row", index=i + 1, calls=child.get("api_calls", "?"), doing=doing) idle = child.get("seconds_since_activity") lines.append(part + (t("gateway.agents.active_ago", seconds=f"{idle:.0f}") if idle is not None else "")) return lines def _usage_agent_stats_lines(agent) -> list[str]: """/usage session block for a live agent: rate limits, token breakdown (matches the CLI), context window and compression count.""" lines: list[str] = [] rl_state = agent.get_rate_limit_state() if rl_state and rl_state.has_data: from agent.rate_limit_tracker import format_rate_limit_compact lines += [t("gateway.usage.rate_limits", state=format_rate_limit_compact(rl_state)), ""] lines += [ t("gateway.usage.header_session"), t("gateway.usage.label_model", model=agent.model), t("gateway.usage.label_input_tokens", count=_fmt(_n(agent, "session_input_tokens"))), t("gateway.usage.label_output_tokens", count=_fmt(_n(agent, "session_output_tokens"))), t("gateway.usage.label_total", count=_fmt(agent.session_total_tokens)), t("gateway.usage.label_api_calls", count=agent.session_api_calls), ] ctx = agent.context_compressor if ctx.last_prompt_tokens > 0: pct = _pct(ctx.last_prompt_tokens, ctx.context_length) from agent.context_breakdown import context_display_source from agent.context_pin import agent_context_suffix mark = "~" if context_display_source(ctx) != "provider_usage" else "" lines.append(t("gateway.usage.label_context", used=mark + _fmt(ctx.last_prompt_tokens), total=_fmt(ctx.context_length), pct=f"{mark}{pct:.0f}") + agent_context_suffix(agent)) if ctx.compression_count: lines.append(t("gateway.usage.label_compressions", count=ctx.compression_count)) return lines def _capped_rows(items: list, render) -> list[str]: """Render up to ``_LIST_CAP`` items via *render* (list of lines each) plus an overflow line.""" lines: list[str] = [] for item in items[:_LIST_CAP]: lines.extend(render(item)) if len(items) > _LIST_CAP: lines.append(t("gateway.agents.more", count=len(items) - _LIST_CAP)) return lines class GatewayStatusCommandsMixin: """Read-only gateway introspection commands: /status, /context, /usage, /agents, /insights, /topup.""" async def _handle_status_command(self, event: MessageEvent) -> str: """Handle /status command.""" from gateway.run import _AGENT_PENDING_SENTINEL source = event.source session_entry = await self.async_session_store.get_or_create_session(source) session_key = session_entry.session_key # Keep the sentinel distinct: a starting/pending run is not a usable agent for # model/context display, but it still occupies the session slot. agent = self._running_agents.get(session_key) is_running = agent is not None and agent is not _AGENT_PENDING_SENTINEL # Pending /queue follow-ups (slot + overflow). adapter = self.adapters.get(source.platform) if source else None queue_depth = self._queue_depth(session_key, adapter=adapter) title, session_row, db_total_tokens, persisted_route = await self._status_session_db_facts( session_entry.session_id ) # Prefer the live or cached agent (actual runtime route + context compressor); fall back # to an active /model override, then SessionDB metadata + last_prompt_tokens so /status # stays useful between turns. Rehydrate first so this precedence survives gateway restarts. status_agent = agent if is_running else self._cached_agent_for(session_key) self._rehydrate_session_model_override(session_key) active_override = self._session_model_override(session_key) or {} model_name, provider_name, context_used, context_total, route = _status_model_route( status_agent, active_override, persisted_route, session_row, session_entry ) if not context_total and model_name: # Same resolver /context uses (off-loop: it can probe /models). A window the resolver only # invented (unknown model → DEFAULT_FALLBACK_CONTEXT) stays hidden rather than being shown # as a real limit; the occupancy-only line below is honest for that case. resolved = await self._resolve_route_context(source, model_name, route) if resolved is not None and resolved.context_source != "default": context_total = _int_value(resolved.context_length) fields = build_status_fields( session_entry.session_id, None, session_row, title=title, model=model_name, provider=provider_name, created=session_entry.created_at, last_activity=session_entry.updated_at, tokens=db_total_tokens, agent_running=is_running, ) lines = [t("gateway.status.header"), "", t("gateway.status.session_id", session_id=fields["session_id"])] if fields["title"]: lines.append(t("gateway.status.title", title=fields["title"])) lines += [t("gateway.status.created", timestamp=fields["created"]), t("gateway.status.last_activity", timestamp=fields["last_activity"])] if fields["model"] and fields["provider"]: lines.append(t("gateway.status.model_provider", model=fields["model"], provider=fields["provider"])) elif fields["model"]: lines.append(t("gateway.status.model", model=fields["model"])) from agent.context_breakdown import context_display_source mark = "~" if context_display_source(getattr(status_agent, "context_compressor", None)) != "provider_usage" else "" if context_total: pct = min(100, round((context_used / context_total) * 100)) from agent.context_pin import agent_context_suffix # (pinned) / (fallback) for the agent's own window compressor_total = _n(getattr(status_agent, "context_compressor", None), "context_length") suffix = agent_context_suffix(status_agent) if compressor_total and compressor_total == context_total else "" lines.append(t("gateway.status.context", used=mark + _fmt(context_used), total=_fmt(context_total), pct=f"{mark}{pct}") + suffix) elif context_used: lines.append(t("gateway.status.context_used", used=mark + _fmt(context_used))) state = t("gateway.status.state_yes") if fields["agent_running"] else t("gateway.status.state_no") lines += [t("gateway.status.tokens", tokens=fields["tokens"]), t("gateway.status.agent_running", state=state)] if queue_depth: lines.append(t("gateway.status.queued", count=queue_depth)) if source.platform == Platform.MATRIX: scope = getattr(self.adapters.get(Platform.MATRIX), "_matrix_session_scope", os.getenv("MATRIX_SESSION_SCOPE", "auto")) lines += [ "", t("gateway.status.matrix_scope_header"), t("gateway.status.matrix_scope_room", room=source.chat_name or source.chat_id), t("gateway.status.matrix_scope_room_id", room_id=source.chat_id), t("gateway.status.matrix_scope_thread", thread_id=source.thread_id or t("gateway.shared.none_value")), t("gateway.status.matrix_scope_mode", scope=scope), t("gateway.status.matrix_scope_key", session_key=self._redact_matrix_session_key(session_key)), ] lines += ["", t("gateway.status.platforms", platforms=', '.join(p.value for p in self.adapters))] return "\n".join(lines) async def _status_session_db_facts(self, session_id: str): """``(title, session_row, db_total_tokens, persisted_route)`` for /status; each fail-open. Token totals come from the SQLite session DB, not SessionStore: run_agent.py persists per-turn token deltas into sessions_db, never into SessionEntry (its total_tokens is always 0). """ db = self._session_db if not db: return None, {}, 0, {} title = await _quiet(lambda: db.get_session_title(session_id)) row = await _quiet(lambda: db.get_session(session_id)) session_row = row if isinstance(row, dict) else {} db_total_tokens = sum( _int_value(session_row.get(k)) for k in ("input_tokens", "output_tokens", "cache_read_tokens", "cache_write_tokens", "reasoning_tokens") ) route = await _quiet(lambda: db.get_recent_session_model_route(session_id)) return title, session_row, db_total_tokens, route if isinstance(route, dict) else {} @staticmethod def _redact_matrix_session_key(session_key: str) -> str: """Return a stable Matrix session-key fingerprint for shared room status.""" digest = hashlib.sha256(str(session_key or "").encode("utf-8")).hexdigest()[:12] return f"sha256:{digest}" async def _handle_context_command(self, event: MessageEvent) -> str: """Handle /context — the deep context-window view (/status has the one-line summary). Gauge, auto-compression threshold/headroom, compression count + last savings, and cumulative throughput (labelled as throughput, NOT context size). Resolution: running agent -> cached agent -> SessionStore/SessionDB metadata -> transcript estimate. ``all`` adds listings. """ source = event.source session_entry = await self.async_session_store.get_or_create_session(source) expanded = event.get_command_args().strip().lower() in {"all", "full", "details"} # Running agent first (mid-turn), then cached agent (between turns). agent = self._resident_agent_for(self._session_key_for_source(source)) or None ctx = getattr(agent, "context_compressor", None) if agent else None used, context_length, model_name, window_suffix = await self._resolve_context_figures( agent, ctx, session_entry, source ) from agent.context_breakdown import context_display_source mark = "~" if context_display_source(ctx) != "provider_usage" else "" # Gauge path: preserve the provenance of the selected occupancy figure. if used > 0 and context_length > 0: pct = _pct(used, context_length) filled = round(pct / 100 * 24) lines = [ t("gateway.context.header"), "", t("gateway.context.model", model=model_name or "?"), t("gateway.context.window", total=_fmt(context_length)) + window_suffix, t("gateway.context.in_use", used=mark + _fmt(used), total=_fmt(context_length), pct=f"{mark}{pct:.0f}"), t("gateway.context.bar", bar="█" * max(0, filled) + "░" * max(0, 24 - filled)), t("gateway.context.headroom", headroom=mark + _fmt(max(0, context_length - used))), "", ] # Full view — compression / throughput need the live agent. lines += _context_compressor_lines(agent, ctx, used) if ctx is not None else [ t("gateway.context.detail_after_first")] # Per-category estimated breakdown (+ optional expanded listings). Same chars/4 engine # the desktop popover and /usage use; plain text (monospace isn't guaranteed on # messaging platforms). Fail-open: rendering errors never break /context. breakdown = await asyncio.to_thread(self._context_breakdown_block, agent, source, expanded) if agent else [] return "\n".join(lines + ([""] + breakdown if breakdown else [])) # Last resort: rough estimate from transcript try: history = await self.async_session_store.load_transcript(session_entry.session_id) except TranscriptReadError: return history_unreadable() if not history: return t("gateway.context.no_data") approx, count = _transcript_estimate(history) return "\n".join([ t("gateway.context.header"), "", t("gateway.context.estimated", count=_fmt(approx), messages=count), t("gateway.context.detail_after_first"), ]) async def _resolve_context_figures(self, agent, ctx, session_entry, source): """``(used, context_length, model_name, window_suffix)`` for /context: used = compressor -> SessionStore; model = agent -> SessionDB row; window = compressor -> gateway model route -> model metadata. The suffix is `` (pinned)`` / `` (fallback)`` from whichever source supplied the window.""" from agent.context_pin import agent_context_suffix, context_window_suffix used = max(0, _n(ctx, "last_prompt_tokens")) or max(0, _int_value(getattr(session_entry, "last_prompt_tokens", 0))) context_length = _n(ctx, "context_length") suffix = agent_context_suffix(agent) if context_length else "" model_name = _clean_str(getattr(agent, "model", "")) if agent is not None else "" if not model_name and self._session_db: row = await _quiet(lambda: self._session_db.get_session(session_entry.session_id)) model_name = _clean_str(row.get("model", "")) if isinstance(row, dict) else "" if not context_length: resolved = await self._resolve_route_context(source, model_name) if resolved is not None: model_name = model_name or resolved.model context_length = _int_value(resolved.context_length) suffix = {"config": " (pinned)", "default": " (fallback)"}.get(resolved.context_source, "") if not context_length and model_name: from agent.model_metadata import get_model_context_length context_length = _int_value( await _quiet(lambda: asyncio.to_thread(get_model_context_length, model_name)) ) suffix = context_window_suffix(context_length, None, model=model_name) return used, context_length, model_name, suffix async def _resolve_route_context(self, source, model_name: str, route: dict | None = None): """``_resolve_gateway_model_context`` for a non-resident session, run off the event loop inside the profile serving ``source`` (multiplex). Fail-open: None on any error.""" from gateway.run import _profile_runtime_scope, _resolve_gateway_model_context def _resolve(): if getattr(getattr(self, "config", None), "multiplex_profiles", False): with _profile_runtime_scope(self._resolve_profile_home_for_source(source)): return _resolve_gateway_model_context(model_name or None, route) return _resolve_gateway_model_context(model_name or None, route) return await _quiet(lambda: asyncio.to_thread(_resolve)) async def _handle_agents_command(self, event: MessageEvent) -> str: """Handle /agents command - list active agents and running tasks.""" from gateway.run import _AGENT_PENDING_SENTINEL from tools.process_registry import format_uptime_short, process_registry now = time.time() current_session_key = self._session_key_for_source(event.source) running_started: dict = getattr(self, "_running_agents_ts", {}) or {} agent_rows: list[dict] = [] for session_key, agent in (getattr(self, "_running_agents", {}) or {}).items(): pending = agent is _AGENT_PENDING_SENTINEL agent_rows.append({ "session_key": session_key, "elapsed": max(0, int(now - float(running_started.get(session_key, now)))), "state": t("gateway.agents.state_starting") if pending else t("gateway.agents.state_running"), "session_id": "" if pending else str(getattr(agent, "session_id", "") or ""), "model": "" if pending else str(getattr(agent, "model", "") or ""), }) agent_rows.sort(key=lambda row: row["elapsed"], reverse=True) procs = _quiet_sync(process_registry.list_sessions, []) running_processes = [p for p in procs if p.get("status") == "running"] background_tasks = [task for task in (getattr(self, "_background_tasks", set()) or set()) if hasattr(task, "done") and not task.done()] # Background (async) delegations — delegate_task(background=true). # Live per-child activity comes from the registry's progress sampler (#51690): api calls, current # tool, seconds since last activity. from tools.async_delegation import list_async_delegations delegations = [d for d in _quiet_sync(list_async_delegations, []) if d.get("status") in ("running", "stalling", "finalizing")] def _agent_row(idx_row): idx, row = idx_row current = t("gateway.agents.this_chat") if row["session_key"] == current_session_key else "" sid = f" · `{row['session_id']}`" if row["session_id"] else "" model = f" · `{row['model']}`" if row["model"] else "" return [f"{idx}. `{row['session_key']}` · {row['state']} · " f"{format_uptime_short(row['elapsed'])}{sid}{model}{current}"] def _proc_row(proc): cmd = _clip(" ".join(str(proc.get("command", "")).split()), 90) return [f"- `{proc.get('session_id', '?')}` · " f"{format_uptime_short(int(proc.get('uptime_seconds', 0)))} · `{cmd}`"] lines = [t("gateway.agents.header"), "", t("gateway.agents.active_agents", count=len(agent_rows))] lines += _capped_rows(list(enumerate(agent_rows, 1)), _agent_row) lines += ["", t("gateway.agents.running_processes", count=len(running_processes))] lines += _capped_rows(running_processes, _proc_row) lines += ["", t("gateway.agents.async_jobs", count=len(background_tasks))] if delegations: lines += ["", t("gateway.agents.background_delegations", count=len(delegations))] lines += _capped_rows(delegations, _agents_delegation_lines) if not (agent_rows or running_processes or background_tasks or delegations): lines += ["", t("gateway.agents.none")] return "\n".join(lines) async def _handle_topup_command(self, event: MessageEvent) -> str: """Handle /topup -- show the Nous balance and hand off to the portal. Does NOT charge, confirm, or track payment (that happens in the browser; the next /topup shows the new balance).""" from agent.account_usage import build_credits_view view = await _quiet(lambda: asyncio.to_thread(build_credits_view, markdown=True)) if view is None or not view.logged_in: return t("gateway.credits.not_logged_in") # Drop the helper's 📈 header; we print our own. lines = [t("gateway.topup.header")] + [ln for ln in view.balance_lines if not ln.lstrip().startswith("📈")] if view.identity_line: lines += ["", view.identity_line] if view.topup_url: lines += ["", t("gateway.topup.portal_link", url=view.topup_url), t("gateway.topup.hint")] return "\n".join(lines) def _context_breakdown_block(self, agent, source, expanded: bool) -> list[str]: """/context per-category block (plain text, chars/4 estimate, same engine as /usage). Runs in a thread; returns [] and never raises.""" try: from agent.context_breakdown import compute_context_details, render_context_breakdown_lines try: payload = self._session_context_breakdown(agent, source) except TranscriptReadError: return [history_unreadable()] # a read failure is not an empty transcript if not (payload.get("categories") or []): return [] details = _quiet_sync(lambda: compute_context_details(agent), {"skills": [], "toolsets": []}) if expanded else None from agent.context_file_sources import context_file_sources_for_agent, render_context_file_lines file_lines = _quiet_sync(lambda: render_context_file_lines(context_file_sources_for_agent(agent)), []) return render_context_breakdown_lines(payload, details=details, grid=False) + ([""] + file_lines if file_lines else []) except Exception: return [] def _session_context_breakdown(self, agent, source) -> dict: """Per-category context estimate (chars/4) for *agent* over the session transcript (sync). Raises ``TranscriptReadError`` (unreadable rows must not pass as an empty transcript).""" from agent.context_breakdown import compute_session_context_breakdown store = self.session_store try: history = store.load_transcript(store.get_or_create_session(source).session_id) or [] except TranscriptReadError: raise except Exception: history = [] return compute_session_context_breakdown(agent, history) def _context_breakdown_lines(self, agent, source) -> list[str]: """/usage per-category context breakdown (chars/4 estimate). Returns [] and never raises.""" try: try: payload = self._session_context_breakdown(agent, source) except TranscriptReadError: return [history_unreadable()] categories = payload.get("categories") or [] if not categories: return [] total = payload.get("estimated_total") or 0 out = [t("gateway.usage.breakdown_header")] for cat in categories: tokens = int(cat.get("tokens") or 0) if tokens <= 0: continue cat_id = str(cat.get("id") or "") label = t(f"gateway.usage.breakdown_cat_{cat_id}") if label.endswith(f"breakdown_cat_{cat_id}"): # missing key: t() echoes it back label = str(cat.get("label") or cat_id) pct = round(tokens / total * 100) if total else 0 out.append(t("gateway.usage.breakdown_line", label=label, count=_fmt(tokens), pct=f"~{pct}")) return out if len(out) > 1 else [] except Exception: return [] async def _handle_usage_command(self, event: MessageEvent) -> str: """Handle /usage -- token usage for the current session (live or cached agent) plus account/credit blocks; ``/usage reset [--force]`` redeems a banked Codex reset credit.""" source = event.source session_key = self._session_key_for_source(source) raw_args = event.get_command_args().strip() args = [a.lower() for a in raw_args.split()] if raw_args else [] wants_reset = bool(args) and args[0] == "reset" if args and not wants_reset: return t("gateway.usage.unknown_subcommand", args=raw_args) # Running agent first (mid-turn), then cached agent (between turns). agent = self._resident_agent_for(session_key) # Provider/base_url/api_key for the account-usage fetch: live agent first, else persisted # billing data on the SessionDB row so `/usage` still returns account info between turns. provider, base_url, api_key = ( getattr(agent, k, None) if agent else None for k in ("provider", "base_url", "api_key") ) if not provider and getattr(self, "_session_db", None) is not None: provider, base_url = await self._persisted_billing_route(source) if not provider: # Fresh or evicted session with no persisted route (e.g. /usage right after login): # fall back to the configured provider, as /status does, so account limits such as # Codex subscription windows still render from on-disk credentials (#15167). provider = await _quiet(lambda: asyncio.to_thread(_configured_provider)) or None if wants_reset: if str(provider or "").strip().lower() != "openai-codex": return t("gateway.usage.reset_wrong_provider") from agent.account_usage import redeem_codex_reset_credit result = await asyncio.to_thread( redeem_codex_reset_credit, base_url=base_url, api_key=api_key, force="--force" in args[1:], ) return result.message # Account usage off the event loop so slow provider APIs don't block the gateway; # failures are non-fatal (account_lines stays []). account_snapshot = provider and await _quiet( lambda: asyncio.to_thread(fetch_account_usage, provider, base_url=base_url, api_key=api_key) ) account_lines = ( render_account_usage_lines(account_snapshot, markdown=True) if account_snapshot else [] ) # Nous credits + monthly-grant gauge (shared with CLI/TUI). Gates on "a Nous account is # logged in" — NOT the inference provider — so a Nous user inferring elsewhere still sees # a balance. Fail-open: never break /usage. from agent.account_usage import nous_credits_lines credits_state = getattr(agent, "_credits_state", None) if agent else None credits_lines = await _quiet( lambda: asyncio.to_thread(nous_credits_lines, markdown=True, credits_state=credits_state), [], ) # Window plans: a chat's /topup has no Keep going control, so point the setting at the Portal. from agent.plan_limits import portal_wording credits_lines = [portal_wording(line) for line in credits_lines] def _with_account_blocks(lines: list[str]) -> str: # Each block is preceded by a blank divider only when something precedes it. for block in (account_lines, credits_lines): if block: if lines: lines.append("") lines.extend(block) return "\n".join(lines) if agent and hasattr(agent, "session_total_tokens") and agent.session_api_calls > 0: lines = _usage_agent_stats_lines(agent) # Per-category breakdown (chars/4 estimate, same engine as the desktop popover): prompt # / tools / skills / memory off the live agent, conversation from the transcript. breakdown_lines = await asyncio.to_thread(self._context_breakdown_lines, agent, source) if breakdown_lines: lines += [""] + breakdown_lines return _with_account_blocks(lines) # No agent at all -- rough count from session history session_entry = await self.async_session_store.get_or_create_session(source) try: history = await self.async_session_store.load_transcript(session_entry.session_id) except TranscriptReadError: return history_unreadable() if history: approx, count = _transcript_estimate(history) return _with_account_blocks([ t("gateway.usage.header_session_info"), t("gateway.usage.label_messages", count=count), t("gateway.usage.label_estimated_context", count=_fmt(approx)), t("gateway.usage.detailed_after_first"), ]) if account_lines or credits_lines: return _with_account_blocks([]) return t("gateway.usage.no_data") async def _persisted_billing_route(self, source): """``(provider, base_url)`` from the SessionDB row / most recent route when no agent is resident.""" async def _rows(): entry = await self.async_session_store.get_or_create_session(source) persisted = await self._session_db.get_session(entry.session_id) or {} route = await self._session_db.get_recent_session_model_route(entry.session_id) return persisted, route if isinstance(route, dict) else {} persisted, recent = await _quiet(_rows, ({}, {})) row = recent if recent.get("billing_provider") else persisted return row.get("billing_provider"), row.get("billing_base_url") async def _handle_insights_command(self, event: MessageEvent) -> str: """Handle /insights [N | --days N] [--source S] -- usage insights and analytics.""" # Normalize Unicode dashes (Telegram/iOS auto-converts -- to em/en dash) args = re.sub(r'[\u2012\u2013\u2014\u2015](days|source)', r'--\1', event.get_command_args().strip()) days, source = 30, None parts = args.split() i = 0 while i < len(parts): flag, value = parts[i], parts[i + 1] if i + 1 < len(parts) else None if flag == "--days" and value is not None: try: days = int(value) except ValueError: return t("gateway.insights.invalid_days", value=value) i += 2 elif flag == "--source" and value is not None: source, i = value, i + 2 else: days = int(flag) if flag.isdigit() else days i += 1 try: from hermes_state_registry import acquire from agent.insights import InsightsEngine def _run_insights(): db = acquire() try: engine = InsightsEngine(db) return engine.format_gateway(engine.generate(days=days, source=source)) finally: from hermes_state_registry import release_or_close release_or_close(db) # Not a bare hop: ``SessionDB()`` resolves ``get_hermes_home()`` at call time, a # contextvar set by ``_profile_runtime_scope``; a default-executor hop starts with an # EMPTY context and would read the DEFAULT profile's state.db. return await self._run_in_executor_with_context(_run_insights) except Exception as e: logger.error("Insights command error: %s", e, exc_info=True) return t("gateway.insights.error", error=e)