cogno-meter 0.1.0__tar.gz

cogno_meter-0.1.0/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2026 Sudoers AI
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

cogno_meter-0.1.0/PKG-INFO

@@ -0,0 +1,110 @@
+Metadata-Version: 2.4
+Name: cogno-meter
+Version: 0.1.0
+Summary: Usage metering + pricing/overage for the Cogno cognitive pipeline — cross-modality (LLM/embedding tokens + STT/TTS chars) billable-token accounting against a plan allowance
+Author: Vinicius Vale
+Maintainer: Sudoers AI
+License-Expression: Apache-2.0
+Project-URL: Homepage, https://github.com/sudoers-ai/cogno-meter
+Project-URL: Repository, https://github.com/sudoers-ai/cogno-meter
+Project-URL: Issues, https://github.com/sudoers-ai/cogno-meter/issues
+Keywords: metering,billing,pricing,tokens,usage,llm,overage,agents
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Office/Business :: Financial
+Classifier: Topic :: Software Development :: Libraries
+Classifier: Operating System :: OS Independent
+Classifier: Typing :: Typed
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Provides-Extra: dev
+Requires-Dist: pytest>=8.0; extra == "dev"
+Requires-Dist: pytest-asyncio>=0.23.0; extra == "dev"
+Requires-Dist: pytest-cov>=5.0; extra == "dev"
+Requires-Dist: ruff>=0.5; extra == "dev"
+Requires-Dist: mypy>=1.8; extra == "dev"
+Dynamic: license-file
+
+# cogno-meter
+
+**Usage metering + pricing/overage for the [Cogno](https://github.com/sudoers-ai/cogno-anima) cognitive pipeline** — cross-modality billable-token accounting against a plan allowance.
+
+`cogno-meter` turns raw usage (LLM/embedding **tokens** + STT/TTS **characters**) into a bill. Pure code, zero dependencies, zero I/O — the host owns the plan values, the price book values, the accumulated-usage persistence, and invoicing. Adapted clean-room from the parent Cogno SaaS's `pricing` / `metering` / `db_billing`.
+
+> Status: **alpha** — the pricing + overage core and the unit suite are in place.
+
+## The billing model
+
+- **Everything counts as billable tokens.** A **local** model costs nothing to run, but its tokens still draw down the monthly allowance.
+- **Audio is char-metered, with a multiplier.** STT/TTS usage is measured in characters and multiplied (default `2×`) so audio is more expensive per the allowance.
+- **Overage is per-plan, not per-model.** Beyond the monthly allowance, the customer is charged the **plan's** rate per million tokens — cheap, because the platform runs local models. **BYOK is not exempt**: over the limit you pay the override even with your own API key.
+- **Provider cost is transparency only.** The real upstream cost (per-model price book) is computed alongside so the client can track external-LLM spend; for external models under BYOK the client pays the provider directly. It is **not** the platform charge.
+
+## Use
+
+```python
+from cogno_meter import meter, Plan, PriceBook, UsageRecord, Modality
+
+plan = Plan(name="Básico", monthly_token_limit=500_000, overage_price=0.05, base_price=20.0)
+book = PriceBook.default()                 # or PriceBook.from_mapping(host_pricing_dict)
+
+records = [                                 # a period's usage (from synapse + vox)
+    UsageRecord(Modality.LLM, "ollama:mistral:latest", tokens_in=1_000_000, tokens_out=200_000),
+    UsageRecord(Modality.TTS, "local:kokoro", chars=50_000),     # 50k × 2 = 100k billable tokens
+]
+
+bill = meter(records, plan=plan, book=book)
+# bill.total_tokens=1_300_000  overage_tokens=800_000  overage_cost=0.04
+# bill.total=20.04  provider_cost_brl=0.0   (all local → free upstream)
+```
+
+External, paid model (transparency cost appears, charge stays plan-based):
+
+```python
+records = [UsageRecord(Modality.LLM, "openai:gpt-4.1-mini", tokens_in=1_000_000, tokens_out=500_000)]
+bill = meter(records, plan=plan, book=book)
+# overage_cost=0.05  total=20.05  (platform charge — plan rate)
+# provider_cost_usd=1.20  provider_cost_brl=6.84  (transparency — client pays OpenAI via BYOK)
+```
+
+## The price book
+
+`PriceBook` resolves a model's rate by exact match → fuzzy version-prefix → provider `_default` (self-hosted = 0), and converts USD→currency. The lib ships an **illustrative** seed (`DEFAULT_RATES`); a host injects its own via `PriceBook.from_mapping(...)` (load your YAML/JSON/DB and pass a plain mapping — no YAML dependency here). Units: LLM/embedding USD per 1M tokens, STT USD per minute, TTS USD per 1M characters.
+
+## Boundary
+
+| Concern | Owner |
+| --- | --- |
+| Price book resolution, per-modality cost, audio char×multiplier, overage math | **cogno-meter** |
+| Plan/allowance/overage-price values, price book values | **host** |
+| Accumulated-usage persistence (the token ledger), invoicing, Stripe | **host** |
+| Per-tenant model overrides, BYOK key storage | **host** |
+
+`meter` consumes usage produced by `cogno-synapse` (tokens) and `cogno-vox` (audio chars); it never imports them — usage is handed in as plain `UsageRecord`s.
+
+## The Cogno ecosystem
+
+`cogno-meter` is one organ of **[Cogno](https://github.com/sudoers-ai)** — a family of
+small, composable, Apache-2.0 libraries that together form a complete
+conversational-agent platform. Each library owns a single concern and stays
+infra-agnostic; a **host** assembles them into a running agent:
+
+![The Cogno ecosystem](docs/assets/cogno-ecosystem.svg)
+
+The open-source libraries are the organs; the **host is the body** that joins
+them. Our reference host — `cogno-host`, with its `cogno-ui` dashboard — is the
+private product layer, but it holds no special powers: everything it does rides
+on the public seams documented in each library's `docs/HOST_INTEGRATION.md`, so
+you can assemble a body of your own.
+
+## Test
+
+```bash
+pip install -e ".[dev]"
+pytest tests/unit -q
+```

cogno_meter-0.1.0/README.md

@@ -0,0 +1,78 @@
+# cogno-meter
+
+**Usage metering + pricing/overage for the [Cogno](https://github.com/sudoers-ai/cogno-anima) cognitive pipeline** — cross-modality billable-token accounting against a plan allowance.
+
+`cogno-meter` turns raw usage (LLM/embedding **tokens** + STT/TTS **characters**) into a bill. Pure code, zero dependencies, zero I/O — the host owns the plan values, the price book values, the accumulated-usage persistence, and invoicing. Adapted clean-room from the parent Cogno SaaS's `pricing` / `metering` / `db_billing`.
+
+> Status: **alpha** — the pricing + overage core and the unit suite are in place.
+
+## The billing model
+
+- **Everything counts as billable tokens.** A **local** model costs nothing to run, but its tokens still draw down the monthly allowance.
+- **Audio is char-metered, with a multiplier.** STT/TTS usage is measured in characters and multiplied (default `2×`) so audio is more expensive per the allowance.
+- **Overage is per-plan, not per-model.** Beyond the monthly allowance, the customer is charged the **plan's** rate per million tokens — cheap, because the platform runs local models. **BYOK is not exempt**: over the limit you pay the override even with your own API key.
+- **Provider cost is transparency only.** The real upstream cost (per-model price book) is computed alongside so the client can track external-LLM spend; for external models under BYOK the client pays the provider directly. It is **not** the platform charge.
+
+## Use
+
+```python
+from cogno_meter import meter, Plan, PriceBook, UsageRecord, Modality
+
+plan = Plan(name="Básico", monthly_token_limit=500_000, overage_price=0.05, base_price=20.0)
+book = PriceBook.default()                 # or PriceBook.from_mapping(host_pricing_dict)
+
+records = [                                 # a period's usage (from synapse + vox)
+    UsageRecord(Modality.LLM, "ollama:mistral:latest", tokens_in=1_000_000, tokens_out=200_000),
+    UsageRecord(Modality.TTS, "local:kokoro", chars=50_000),     # 50k × 2 = 100k billable tokens
+]
+
+bill = meter(records, plan=plan, book=book)
+# bill.total_tokens=1_300_000  overage_tokens=800_000  overage_cost=0.04
+# bill.total=20.04  provider_cost_brl=0.0   (all local → free upstream)
+```
+
+External, paid model (transparency cost appears, charge stays plan-based):
+
+```python
+records = [UsageRecord(Modality.LLM, "openai:gpt-4.1-mini", tokens_in=1_000_000, tokens_out=500_000)]
+bill = meter(records, plan=plan, book=book)
+# overage_cost=0.05  total=20.05  (platform charge — plan rate)
+# provider_cost_usd=1.20  provider_cost_brl=6.84  (transparency — client pays OpenAI via BYOK)
+```
+
+## The price book
+
+`PriceBook` resolves a model's rate by exact match → fuzzy version-prefix → provider `_default` (self-hosted = 0), and converts USD→currency. The lib ships an **illustrative** seed (`DEFAULT_RATES`); a host injects its own via `PriceBook.from_mapping(...)` (load your YAML/JSON/DB and pass a plain mapping — no YAML dependency here). Units: LLM/embedding USD per 1M tokens, STT USD per minute, TTS USD per 1M characters.
+
+## Boundary
+
+| Concern | Owner |
+| --- | --- |
+| Price book resolution, per-modality cost, audio char×multiplier, overage math | **cogno-meter** |
+| Plan/allowance/overage-price values, price book values | **host** |
+| Accumulated-usage persistence (the token ledger), invoicing, Stripe | **host** |
+| Per-tenant model overrides, BYOK key storage | **host** |
+
+`meter` consumes usage produced by `cogno-synapse` (tokens) and `cogno-vox` (audio chars); it never imports them — usage is handed in as plain `UsageRecord`s.
+
+## The Cogno ecosystem
+
+`cogno-meter` is one organ of **[Cogno](https://github.com/sudoers-ai)** — a family of
+small, composable, Apache-2.0 libraries that together form a complete
+conversational-agent platform. Each library owns a single concern and stays
+infra-agnostic; a **host** assembles them into a running agent:
+
+![The Cogno ecosystem](docs/assets/cogno-ecosystem.svg)
+
+The open-source libraries are the organs; the **host is the body** that joins
+them. Our reference host — `cogno-host`, with its `cogno-ui` dashboard — is the
+private product layer, but it holds no special powers: everything it does rides
+on the public seams documented in each library's `docs/HOST_INTEGRATION.md`, so
+you can assemble a body of your own.
+
+## Test
+
+```bash
+pip install -e ".[dev]"
+pytest tests/unit -q
+```

cogno_meter-0.1.0/cogno_meter/__init__.py

@@ -0,0 +1,66 @@
+"""
+cogno-meter — usage metering + pricing/overage for the Cogno stack.
+
+Cross-modality usage (LLM/embedding tokens + STT/TTS characters) is normalized to
+**billable tokens** and billed against a plan: a monthly token allowance plus a
+per-million **overage** at the plan's rate. Local models cost nothing to run but
+still consume the allowance; audio is char-metered with a multiplier so it costs
+more. The real upstream **provider cost** is computed alongside, as transparency
+only (with BYOK the client pays the provider directly).
+
+Pure code, zero dependencies, zero I/O: the host owns the plan values, the price
+book values, the accumulated-usage persistence, and invoicing. Adapted from the
+parent cogno's pricing/metering/billing.
+"""
+
+from importlib.metadata import PackageNotFoundError
+from importlib.metadata import version as _dist_version
+
+try:
+    __version__ = _dist_version("cogno-meter")
+except PackageNotFoundError:  # source tree without an installed dist (e.g. vendored checkout)
+    __version__ = "0.0.0"
+
+
+from cogno_meter.types import Bill, Modality, Plan, UsageRecord
+from cogno_meter.pricing import (
+    DEFAULT_AUDIO_MULTIPLIER,
+    DEFAULT_RATES,
+    DEFAULT_USD_BRL_RATE,
+    PriceBook,
+)
+from cogno_meter.billing import (
+    compute_bill,
+    meter,
+    total_billable_tokens,
+    total_provider_cost_usd,
+)
+from cogno_meter.budget import (
+    BLOCKING_STATUSES,
+    WARN_THRESHOLD,
+    BudgetInputs,
+    BudgetReason,
+    BudgetVerdict,
+    evaluate_budget,
+)
+
+__all__ = [
+    "Modality",
+    "UsageRecord",
+    "Plan",
+    "Bill",
+    "PriceBook",
+    "DEFAULT_RATES",
+    "DEFAULT_USD_BRL_RATE",
+    "DEFAULT_AUDIO_MULTIPLIER",
+    "meter",
+    "compute_bill",
+    "total_billable_tokens",
+    "total_provider_cost_usd",
+    "evaluate_budget",
+    "BudgetInputs",
+    "BudgetVerdict",
+    "BudgetReason",
+    "BLOCKING_STATUSES",
+    "WARN_THRESHOLD",
+]

cogno_meter-0.1.0/cogno_meter/billing.py

@@ -0,0 +1,71 @@
+"""
+cogno_meter.billing — aggregate usage and compute the bill.
+
+The overage math is ported from the parent ``cogno/core/db_billing.py`` (the
+pure calculation, not the DB/Stripe layer): everything counts toward the monthly
+token total (local models included, at zero provider cost); the customer is
+charged the **plan's** per-million ``overage_price`` on tokens beyond the
+allowance — independent of whose API key ran the model (BYOK is not exempt).
+
+The provider cost (price book) is aggregated alongside, purely as transparency.
+"""
+
+from __future__ import annotations
+
+import logging
+from typing import Iterable
+
+from cogno_meter.pricing import PriceBook
+from cogno_meter.types import Bill, Plan, UsageRecord
+
+# Pure functions — DEBUG-only (the caller logs the returned Bill; the
+# budget-block decision belongs to the host). See LOGGING.md.
+logger = logging.getLogger("cogno_meter.billing")
+
+
+def total_billable_tokens(records: Iterable[UsageRecord], book: PriceBook) -> int:
+    return sum(book.billable_tokens(r) for r in records)
+
+
+def total_provider_cost_usd(records: Iterable[UsageRecord], book: PriceBook) -> float:
+    return sum(book.usage_cost_usd(r) for r in records)
+
+
+def compute_bill(
+    *,
+    plan: Plan,
+    billable_tokens: int,
+    provider_cost_usd: float,
+    book: PriceBook,
+) -> Bill:
+    """Apply the plan's allowance + per-million overage to a period's totals."""
+    overage_tokens = max(0, billable_tokens - plan.monthly_token_limit)
+    overage_cost = (overage_tokens / 1_000_000) * plan.overage_price
+    total = plan.base_price + overage_cost
+    logger.debug(
+        "event=compute_bill total_tokens=%d allowance=%d overage_tokens=%d "
+        "overage_cost=%.6f total=%.6f provider_cost_usd=%.6f",
+        billable_tokens, plan.monthly_token_limit, overage_tokens,
+        overage_cost, total, provider_cost_usd,
+    )
+    return Bill(
+        total_tokens=billable_tokens,
+        allowance=plan.monthly_token_limit,
+        overage_tokens=overage_tokens,
+        overage_cost=round(overage_cost, 6),
+        base_price=plan.base_price,
+        total=round(total, 6),
+        provider_cost_usd=round(provider_cost_usd, 6),
+        provider_cost_brl=round(provider_cost_usd * book.usd_brl_rate, 6),
+    )
+
+
+def meter(records: Iterable[UsageRecord], *, plan: Plan, book: PriceBook) -> Bill:
+    """End-to-end: aggregate a period's usage records into a :class:`Bill`."""
+    records = list(records)
+    return compute_bill(
+        plan=plan,
+        billable_tokens=total_billable_tokens(records, book),
+        provider_cost_usd=total_provider_cost_usd(records, book),
+        book=book,
+    )

cogno_meter-0.1.0/cogno_meter/budget.py

@@ -0,0 +1,106 @@
+"""Budget/quota decision — the pure gate the host wires to its stores.
+
+The parent's ``BudgetGuard`` (``cogno/core/budget_guard.py``) mixed store I/O, currency, dates and
+pt-BR messages into one class. Here the **decision** is extracted as pure policy: the host fetches
+the numbers (daily spend, monthly usage, subscription state), converts them to the plan currency,
+parses expiry, and passes them in; this module applies the layered budget policy and returns a
+:class:`BudgetVerdict`. Zero I/O, no dates, no FX, no locale — a blocked verdict names the
+:class:`BudgetReason`, and the host renders the localized message.
+
+Layers (order): tenant daily budget → identity daily budget → subscription status/expiry →
+monthly token quota (hard-stop ONLY when the plan has no overage). Each daily layer warns at 80%.
+A budget of ``None`` / ``<= 0`` is *unset* (no limit) — never blocks. Never raises.
+"""
+
+from __future__ import annotations
+
+from dataclasses import dataclass, field
+from enum import Enum
+from typing import List, Optional
+
+# Subscription states that block a turn (an active or unset/"inactive" free plan runs). The host
+# may override via ``BudgetInputs.blocking_statuses``.
+BLOCKING_STATUSES: "frozenset[str]" = frozenset({"canceled", "past_due", "suspended"})
+# Soft-warning threshold (fraction of a budget/quota that raises a warning, not a block).
+WARN_THRESHOLD = 0.8
+
+
+class BudgetReason(str, Enum):
+    """Why a turn was blocked — the host maps this to a localized user message."""
+
+    TENANT_DAILY = "tenant_daily_budget"
+    IDENTITY_DAILY = "identity_daily_budget"
+    SUBSCRIPTION_INACTIVE = "subscription_inactive"
+    SUBSCRIPTION_EXPIRED = "subscription_expired"
+    MONTHLY_TOKENS = "monthly_tokens"
+
+
+@dataclass(frozen=True)
+class BudgetInputs:
+    """Everything pre-fetched + normalized by the host. Spends are in the SAME unit as the budgets
+    (the host applies any FX before building this); ``subscription_expired`` is pre-computed (the
+    host parses the date). A ``None``/non-positive budget or a 0 monthly limit means *unset*."""
+
+    tenant_daily_budget: Optional[float] = None
+    tenant_daily_spend: float = 0.0
+    identity_daily_budget: Optional[float] = None
+    identity_daily_spend: float = 0.0
+    subscription_status: Optional[str] = None    # None → no subscription (skip the layer)
+    subscription_expired: bool = False
+    monthly_token_limit: int = 0
+    monthly_tokens_used: int = 0
+    overage_price: float = 0.0                    # >0 → paid plan keeps running over the limit
+    plan_limits: dict = field(default_factory=dict)   # loop-depth caps to pass through
+    call_limit: Optional[float] = None                # per-call ceiling to pass through
+    blocking_statuses: "frozenset[str]" = BLOCKING_STATUSES
+
+
+@dataclass
+class BudgetVerdict:
+    """The decision. ``plan_limits``/``call_limit`` pass through for the host to apply on an
+    allowed turn; ``warnings`` carries 80%-threshold notices (``"<reason>:80pct"``)."""
+
+    blocked: bool = False
+    reason: Optional[BudgetReason] = None
+    plan_limits: dict = field(default_factory=dict)
+    call_limit: Optional[float] = None
+    warnings: List[str] = field(default_factory=list)
+
+
+def evaluate_budget(inp: BudgetInputs) -> BudgetVerdict:
+    """Apply the layered budget policy → a :class:`BudgetVerdict`. Currency/locale/time-agnostic."""
+    warnings: List[str] = []
+
+    def _block(reason: BudgetReason) -> BudgetVerdict:
+        return BudgetVerdict(blocked=True, reason=reason, plan_limits=inp.plan_limits,
+                             call_limit=inp.call_limit, warnings=warnings)
+
+    # 1) tenant daily budget, then 2) identity daily budget — block at 100%, warn at 80%.
+    for budget, spend, reason in (
+        (inp.tenant_daily_budget, inp.tenant_daily_spend, BudgetReason.TENANT_DAILY),
+        (inp.identity_daily_budget, inp.identity_daily_spend, BudgetReason.IDENTITY_DAILY),
+    ):
+        if budget is None or budget <= 0:   # unset → no limit (narrows budget to float below)
+            continue
+        if spend >= budget:
+            return _block(reason)
+        if spend >= budget * WARN_THRESHOLD:
+            warnings.append(f"{reason.value}:80pct")
+
+    # 3) subscription status / expiry (only when a subscription is present)
+    if inp.subscription_status is not None:
+        if inp.subscription_status.lower() in inp.blocking_statuses:
+            return _block(BudgetReason.SUBSCRIPTION_INACTIVE)
+        if inp.subscription_expired:
+            return _block(BudgetReason.SUBSCRIPTION_EXPIRED)
+
+    # 4) monthly token quota — hard-stop ONLY for a plan with no overage (Free); paid plans run on
+    if inp.monthly_token_limit > 0:
+        used, limit = inp.monthly_tokens_used, inp.monthly_token_limit
+        if used >= limit and inp.overage_price <= 0:
+            return _block(BudgetReason.MONTHLY_TOKENS)
+        if used >= limit * WARN_THRESHOLD:
+            warnings.append("monthly_tokens:80pct")
+
+    return BudgetVerdict(blocked=False, reason=None, plan_limits=inp.plan_limits,
+                         call_limit=inp.call_limit, warnings=warnings)

cogno_meter-0.1.0/cogno_meter/pricing.py

@@ -0,0 +1,211 @@
+"""
+cogno_meter.pricing — the price book + per-modality cost resolution.
+
+Adapted clean-room from the parent ``cogno/core/pricing.py``: model rates with
+exact → fuzzy-prefix (version suffixes) → ``_default`` resolution, USD→currency
+conversion, and per-modality cost. Two distinct numbers come out of here:
+
+  * **billable_tokens** — what counts toward the plan allowance/overage (LLM &
+    embedding tokens directly; audio = ``chars × audio_multiplier``).
+  * **cost_usd** — the real upstream provider cost (transparency only). Local /
+    self-hosted models resolve to ``_default = 0``.
+
+The lib ships a default seed (illustrative rates); a host injects its own via
+``PriceBook.from_mapping(...)``. No YAML dependency — the host loads its config
+(YAML/JSON/DB) and hands over a plain mapping.
+"""
+
+from __future__ import annotations
+
+import copy
+import logging
+from dataclasses import dataclass, field
+
+from cogno_meter.types import Modality, UsageRecord
+
+# Pure functions: no WARNING/INFO (the budget-block decision is the host's, and
+# the caller logs the returned Bill). DEBUG only — inspect the calculation.
+logger = logging.getLogger("cogno_meter.pricing")
+
+# Illustrative seed — values are examples (verify against providers; host overrides).
+# llm: USD per 1M tokens (input/output). embedding: USD per 1M tokens.
+# stt: USD per minute of audio. tts: USD per 1M characters. _default: self-hosted = 0.
+DEFAULT_RATES: dict = {
+    "llm": {
+        # exact keys for every catalogued model — the fuzzy resolver would otherwise let a
+        # bare "gpt-5-mini" greedily prefix-match "gpt-5"'s far pricier rate.
+        "openai:gpt-5-nano": {"input": 0.05, "output": 0.40},
+        "openai:gpt-4.1-nano": {"input": 0.10, "output": 0.40},
+        "openai:gpt-4o-mini": {"input": 0.15, "output": 0.60},
+        "openai:gpt-5.4-nano": {"input": 0.20, "output": 1.25},
+        "openai:gpt-5-mini": {"input": 0.25, "output": 2.00},
+        "openai:gpt-4.1-mini": {"input": 0.40, "output": 1.60},
+        "openai:gpt-5.4-mini": {"input": 0.75, "output": 4.50},
+        "openai:gpt-5.6-luna": {"input": 1.00, "output": 6.00},
+        "openai:gpt-5": {"input": 1.25, "output": 10.00},
+        "openai:gpt-4.1": {"input": 2.00, "output": 8.00},
+        "openai:gpt-5.6-terra": {"input": 2.50, "output": 15.00},
+        "openai:gpt-5.6-sol": {"input": 5.00, "output": 30.00},
+        "openai:gpt-5.5-pro": {"input": 30.00, "output": 180.00},
+        "openai:gpt-4o": {"input": 2.50, "output": 10.00},
+        # anthropic — real per-MTok rates (Opus 4.6=$5/$25, Sonnet 4.5=$3/$15, Haiku 4.5=$1/$5);
+        # the original Opus 4 keeps its launch $15/$75.
+        # Anthropic API IDs are hyphenated (claude-opus-4-6, not …4.6) — must match the id the
+        # backend reports into the ledger, or the rate resolves to _default (0).
+        "anthropic:claude-haiku-4-5": {"input": 1.00, "output": 5.00},
+        "anthropic:claude-sonnet-4-0": {"input": 3.00, "output": 15.00},
+        "anthropic:claude-sonnet-4-5": {"input": 3.00, "output": 15.00},
+        "anthropic:claude-opus-4-0": {"input": 15.00, "output": 75.00},
+        "anthropic:claude-opus-4-5": {"input": 5.00, "output": 25.00},
+        "anthropic:claude-opus-4-6": {"input": 5.00, "output": 25.00},
+        # gemini — 2.5 family verified vs ai.google.dev; 3.x preview partly estimated.
+        "gemini:gemini-2.5-flash-lite": {"input": 0.10, "output": 0.40},   # verified
+        "gemini:gemini-2.5-flash": {"input": 0.30, "output": 2.50},        # verified
+        "gemini:gemini-3-flash": {"input": 0.50, "output": 3.00},          # input verified, output est.
+        "gemini:gemini-2.5-pro": {"input": 1.25, "output": 10.00},         # verified (≤200k prompt)
+        "gemini:gemini-3-pro": {"input": 2.00, "output": 12.00},           # est. (aligned to 3.1-pro)
+        "gemini:gemini-3.1-pro": {"input": 2.00, "output": 12.00},         # verified (3.1 Pro Preview)
+        # grok — verified vs xAI pricing; grok-3-mini estimated. (grok-4.20 was fictional →
+        # dropped; real flagship is grok-4.5.)
+        "grok:grok-3-mini": {"input": 0.30, "output": 0.50},               # est.
+        "grok:grok-3": {"input": 2.00, "output": 10.00},                   # verified (legacy)
+        "grok:grok-4": {"input": 3.00, "output": 15.00},                   # verified
+        "grok:grok-4.1-fast": {"input": 0.20, "output": 0.50},             # verified
+        "grok:grok-4.3": {"input": 1.25, "output": 2.50},                  # verified
+        "grok:grok-4.5": {"input": 2.00, "output": 6.00},                  # verified (flagship)
+        "ollama:_default": {"input": 0.0, "output": 0.0},  # self-hosted
+        "_default": {"input": 0.0, "output": 0.0},
+    },
+    "embedding": {
+        "openai:text-embedding-3-small": 0.02,
+        "openai:text-embedding-3-large": 0.13,
+        "_default": 0.0,
+    },
+    "stt": {  # USD per minute
+        "openai:whisper-1": 0.006,
+        "groq:whisper-large-v3-turbo": 0.04,
+        "_default": 0.0,  # self-hosted faster-whisper
+    },
+    "tts": {  # USD per 1M characters
+        "openai:tts-1": 15.00,
+        "openai:tts-1-hd": 30.00,
+        "xai:grok-2-tts": 4.20,
+        "_default": 0.0,  # self-hosted Kokoro
+    },
+}
+
+DEFAULT_USD_BRL_RATE = 5.70
+DEFAULT_AUDIO_MULTIPLIER = 2.0
+
+
+@dataclass
+class PriceBook:
+    rates: dict = field(default_factory=lambda: copy.deepcopy(DEFAULT_RATES))
+    usd_brl_rate: float = DEFAULT_USD_BRL_RATE
+    audio_multiplier: float = DEFAULT_AUDIO_MULTIPLIER  # chars × this → billable tokens
+
+    @classmethod
+    def default(cls) -> "PriceBook":
+        return cls()
+
+    @classmethod
+    def from_mapping(cls, mapping: dict) -> "PriceBook":
+        """Build from a host-supplied mapping (the parsed pricing config)."""
+        return cls(
+            rates={k: v for k, v in mapping.items()
+                   if k in ("llm", "embedding", "stt", "tts")},
+            usd_brl_rate=float(mapping.get("usd_brl_rate", DEFAULT_USD_BRL_RATE)),
+            audio_multiplier=float(mapping.get("audio_multiplier", DEFAULT_AUDIO_MULTIPLIER)),
+        )
+
+    # ── rate resolution: exact → prefix-fuzzy → BARE (no 'provider:' prefix) → _default ──
+    #
+    # The ledger stores the backend's bare model name ('gpt-4o', 'qwen3:8b'), but the rate keys
+    # carry a 'provider:' prefix ('openai:gpt-4o'). So besides matching a prefixed model, also
+    # match a bare model against the key's model part (exact first so 'gpt-4o' doesn't greedily
+    # grab 'gpt-4o-mini', then longest-prefix fuzzy for versioned names like 'gpt-4o-mini-2024…').
+    @staticmethod
+    def _bare(key: str) -> str:
+        return key.split(":", 1)[1] if ":" in key else key
+
+    @staticmethod
+    def _resolve(table: dict, model: str):
+        if model in table:
+            logger.debug("event=rate_resolve model=%s match=exact", model)
+            return table[model]
+        items = [(k, r) for k, r in table.items() if k != "_default"]
+        # Longest-key-first so a short key that string-prefixes a longer one (``gpt-5`` vs
+        # ``gpt-5.5-pro``) never steals the more specific model's rate. All fuzzy passes below
+        # must be longest-first; step 2 used to iterate in dict-insertion order and mispriced
+        # dated builds of flagship models (~19x under-report on the cost-transparency figure).
+        prefix_items = sorted(items, key=lambda kv: len(kv[0]), reverse=True)
+        for key, rate in prefix_items:                # prefixed model vs prefixed key (versioned)
+            if model.startswith(key):
+                logger.debug("event=rate_resolve model=%s match=fuzzy key=%s", model, key)
+                return rate
+        for key, rate in items:                       # bare model == the key's model part
+            if model == PriceBook._bare(key):
+                logger.debug("event=rate_resolve model=%s match=bare key=%s", model, key)
+                return rate
+        for key, rate in sorted(items, key=lambda kv: len(PriceBook._bare(kv[0])), reverse=True):
+            bare = PriceBook._bare(key)               # bare versioned → longest bare prefix wins
+            if bare and model.startswith(bare):
+                logger.debug("event=rate_resolve model=%s match=bare_fuzzy key=%s", model, key)
+                return rate
+        # Per-provider catch-all (``openai:_default``) before the global one, so a host can set a
+        # non-zero default per provider instead of every un-catalogued model silently costing 0.
+        if ":" in model:
+            provider_default = table.get(f"{model.split(':', 1)[0]}:_default")
+            if provider_default is not None:
+                logger.debug("event=rate_resolve model=%s match=provider_default", model)
+                return provider_default
+        logger.debug("event=rate_resolve model=%s match=default", model)
+        return table.get("_default")
+
+    # ── provider cost (transparency), in USD ──────────────────────────
+    def llm_cost_usd(self, model: str, tokens_in: int, tokens_out: int) -> float:
+        rates = self._resolve(self.rates.get("llm", {}), model)
+        if not isinstance(rates, dict):
+            return 0.0
+        return (tokens_in / 1_000_000) * float(rates.get("input", 0)) + \
+               (tokens_out / 1_000_000) * float(rates.get("output", 0))
+
+    def embedding_cost_usd(self, model: str, tokens: int) -> float:
+        rate = self._resolve(self.rates.get("embedding", {}), model)
+        return (tokens / 1_000_000) * float(rate or 0.0)
+
+    def stt_cost_usd(self, model: str, minutes: float) -> float:
+        rate = self._resolve(self.rates.get("stt", {}), model)
+        return float(minutes) * float(rate or 0.0)
+
+    def tts_cost_usd(self, model: str, chars: int) -> float:
+        rate = self._resolve(self.rates.get("tts", {}), model)
+        return (chars / 1_000_000) * float(rate or 0.0)
+
+    def usage_cost_usd(self, rec: UsageRecord) -> float:
+        if rec.modality == Modality.LLM:
+            return self.llm_cost_usd(rec.model, int(rec.tokens_in or 0), int(rec.tokens_out or 0))
+        if rec.modality == Modality.EMBEDDING:
+            return self.embedding_cost_usd(rec.model, int(rec.tokens_in or 0) or int(rec.tokens_out or 0))
+        if rec.modality == Modality.STT:
+            return self.stt_cost_usd(rec.model, rec.minutes)
+        if rec.modality == Modality.TTS:
+            return self.tts_cost_usd(rec.model, rec.chars)
+        return 0.0
+
+    # ── billable tokens (toward the allowance/overage) ────────────────
+    def billable_tokens(self, rec: UsageRecord) -> int:
+        """Tokens that count toward the monthly allowance. LLM/embedding use the
+        token counts directly; audio is metered by chars × ``audio_multiplier``.
+
+        Token fields are coerced (``None``/missing → 0): one malformed ledger record must not
+        raise ``TypeError`` and abort billing for the whole period."""
+        tin, tout = int(rec.tokens_in or 0), int(rec.tokens_out or 0)
+        if rec.modality == Modality.LLM:
+            return tin + tout
+        # Embedding: mirror the cost path (``tokens_in or tokens_out``) so billable tokens and
+        # the priced amount interpret the same record identically — embeddings carry input only.
+        if rec.modality == Modality.EMBEDDING:
+            return tin or tout
+        # STT / TTS — always char-metered, scaled up by the audio multiplier.
+        return int(round((rec.chars or 0) * self.audio_multiplier))

cogno_meter-0.1.0/cogno_meter/types.py

@@ -0,0 +1,64 @@
+"""
+cogno_meter.types — usage records, plans, and the computed bill.
+
+All cross-modality usage is normalized to **billable tokens** (the unit the
+monthly allowance and overage are denominated in). Audio (STT/TTS) is metered by
+**characters**, multiplied by a configurable factor so it draws down the
+allowance faster (see ``PriceBook.audio_multiplier``).
+"""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+from enum import Enum
+
+
+class Modality(str, Enum):
+    LLM = "llm"
+    EMBEDDING = "embedding"
+    STT = "stt"
+    TTS = "tts"
+
+
+@dataclass
+class UsageRecord:
+    """One unit of metered usage (a stage call, a TTS synthesis, …).
+
+    LLM/embedding usage carries token counts; audio (STT/TTS) carries ``chars``
+    (always char-metered) and, optionally, ``minutes`` of audio for the
+    provider-cost transparency figure (STT is priced per minute upstream).
+    """
+
+    modality: Modality
+    model: str                 # "provider:model" (the model that actually ran)
+    tokens_in: int = 0
+    tokens_out: int = 0
+    chars: int = 0             # audio: STT transcribed / TTS input characters
+    minutes: float = 0.0       # audio duration (optional; STT provider-cost only)
+
+
+@dataclass
+class Plan:
+    """A tenant's plan. Values are host-owned (Stripe/DB) — these are just the
+    parameters the overage math needs."""
+
+    name: str
+    monthly_token_limit: int   # free billable tokens included per period
+    overage_price: float       # charge per 1M tokens over the limit (plan currency)
+    base_price: float = 0.0    # subscription base for the period (plan currency)
+
+
+@dataclass
+class Bill:
+    """The computed bill for a period. ``total`` is what the platform charges the
+    customer; ``provider_cost_*`` is transparency only (real upstream cost, paid
+    by the client via BYOK for external models — NOT the platform charge)."""
+
+    total_tokens: int          # billable tokens (incl. audio chars × multiplier)
+    allowance: int
+    overage_tokens: int
+    overage_cost: float        # plan currency
+    base_price: float
+    total: float               # base_price + overage_cost (plan currency)
+    provider_cost_usd: float   # transparency
+    provider_cost_brl: float   # transparency (usd × usd_brl_rate)

cogno_meter-0.1.0/cogno_meter.egg-info/PKG-INFO

@@ -0,0 +1,110 @@
+Metadata-Version: 2.4
+Name: cogno-meter
+Version: 0.1.0
+Summary: Usage metering + pricing/overage for the Cogno cognitive pipeline — cross-modality (LLM/embedding tokens + STT/TTS chars) billable-token accounting against a plan allowance
+Author: Vinicius Vale
+Maintainer: Sudoers AI
+License-Expression: Apache-2.0
+Project-URL: Homepage, https://github.com/sudoers-ai/cogno-meter
+Project-URL: Repository, https://github.com/sudoers-ai/cogno-meter
+Project-URL: Issues, https://github.com/sudoers-ai/cogno-meter/issues
+Keywords: metering,billing,pricing,tokens,usage,llm,overage,agents
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Office/Business :: Financial
+Classifier: Topic :: Software Development :: Libraries
+Classifier: Operating System :: OS Independent
+Classifier: Typing :: Typed
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Provides-Extra: dev
+Requires-Dist: pytest>=8.0; extra == "dev"
+Requires-Dist: pytest-asyncio>=0.23.0; extra == "dev"
+Requires-Dist: pytest-cov>=5.0; extra == "dev"
+Requires-Dist: ruff>=0.5; extra == "dev"
+Requires-Dist: mypy>=1.8; extra == "dev"
+Dynamic: license-file
+
+# cogno-meter
+
+**Usage metering + pricing/overage for the [Cogno](https://github.com/sudoers-ai/cogno-anima) cognitive pipeline** — cross-modality billable-token accounting against a plan allowance.
+
+`cogno-meter` turns raw usage (LLM/embedding **tokens** + STT/TTS **characters**) into a bill. Pure code, zero dependencies, zero I/O — the host owns the plan values, the price book values, the accumulated-usage persistence, and invoicing. Adapted clean-room from the parent Cogno SaaS's `pricing` / `metering` / `db_billing`.
+
+> Status: **alpha** — the pricing + overage core and the unit suite are in place.
+
+## The billing model
+
+- **Everything counts as billable tokens.** A **local** model costs nothing to run, but its tokens still draw down the monthly allowance.
+- **Audio is char-metered, with a multiplier.** STT/TTS usage is measured in characters and multiplied (default `2×`) so audio is more expensive per the allowance.
+- **Overage is per-plan, not per-model.** Beyond the monthly allowance, the customer is charged the **plan's** rate per million tokens — cheap, because the platform runs local models. **BYOK is not exempt**: over the limit you pay the override even with your own API key.
+- **Provider cost is transparency only.** The real upstream cost (per-model price book) is computed alongside so the client can track external-LLM spend; for external models under BYOK the client pays the provider directly. It is **not** the platform charge.
+
+## Use
+
+```python
+from cogno_meter import meter, Plan, PriceBook, UsageRecord, Modality
+
+plan = Plan(name="Básico", monthly_token_limit=500_000, overage_price=0.05, base_price=20.0)
+book = PriceBook.default()                 # or PriceBook.from_mapping(host_pricing_dict)
+
+records = [                                 # a period's usage (from synapse + vox)
+    UsageRecord(Modality.LLM, "ollama:mistral:latest", tokens_in=1_000_000, tokens_out=200_000),
+    UsageRecord(Modality.TTS, "local:kokoro", chars=50_000),     # 50k × 2 = 100k billable tokens
+]
+
+bill = meter(records, plan=plan, book=book)
+# bill.total_tokens=1_300_000  overage_tokens=800_000  overage_cost=0.04
+# bill.total=20.04  provider_cost_brl=0.0   (all local → free upstream)
+```
+
+External, paid model (transparency cost appears, charge stays plan-based):
+
+```python
+records = [UsageRecord(Modality.LLM, "openai:gpt-4.1-mini", tokens_in=1_000_000, tokens_out=500_000)]
+bill = meter(records, plan=plan, book=book)
+# overage_cost=0.05  total=20.05  (platform charge — plan rate)
+# provider_cost_usd=1.20  provider_cost_brl=6.84  (transparency — client pays OpenAI via BYOK)
+```
+
+## The price book
+
+`PriceBook` resolves a model's rate by exact match → fuzzy version-prefix → provider `_default` (self-hosted = 0), and converts USD→currency. The lib ships an **illustrative** seed (`DEFAULT_RATES`); a host injects its own via `PriceBook.from_mapping(...)` (load your YAML/JSON/DB and pass a plain mapping — no YAML dependency here). Units: LLM/embedding USD per 1M tokens, STT USD per minute, TTS USD per 1M characters.
+
+## Boundary
+
+| Concern | Owner |
+| --- | --- |
+| Price book resolution, per-modality cost, audio char×multiplier, overage math | **cogno-meter** |
+| Plan/allowance/overage-price values, price book values | **host** |
+| Accumulated-usage persistence (the token ledger), invoicing, Stripe | **host** |
+| Per-tenant model overrides, BYOK key storage | **host** |
+
+`meter` consumes usage produced by `cogno-synapse` (tokens) and `cogno-vox` (audio chars); it never imports them — usage is handed in as plain `UsageRecord`s.
+
+## The Cogno ecosystem
+
+`cogno-meter` is one organ of **[Cogno](https://github.com/sudoers-ai)** — a family of
+small, composable, Apache-2.0 libraries that together form a complete
+conversational-agent platform. Each library owns a single concern and stays
+infra-agnostic; a **host** assembles them into a running agent:
+
+![The Cogno ecosystem](docs/assets/cogno-ecosystem.svg)
+
+The open-source libraries are the organs; the **host is the body** that joins
+them. Our reference host — `cogno-host`, with its `cogno-ui` dashboard — is the
+private product layer, but it holds no special powers: everything it does rides
+on the public seams documented in each library's `docs/HOST_INTEGRATION.md`, so
+you can assemble a body of your own.
+
+## Test
+
+```bash
+pip install -e ".[dev]"
+pytest tests/unit -q
+```

cogno_meter-0.1.0/cogno_meter.egg-info/SOURCES.txt

@@ -0,0 +1,14 @@
+LICENSE
+README.md
+pyproject.toml
+cogno_meter/__init__.py
+cogno_meter/billing.py
+cogno_meter/budget.py
+cogno_meter/pricing.py
+cogno_meter/py.typed
+cogno_meter/types.py
+cogno_meter.egg-info/PKG-INFO
+cogno_meter.egg-info/SOURCES.txt
+cogno_meter.egg-info/dependency_links.txt
+cogno_meter.egg-info/requires.txt
+cogno_meter.egg-info/top_level.txt

cogno_meter-0.1.0/cogno_meter.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

cogno_meter-0.1.0/cogno_meter.egg-info/requires.txt

@@ -0,0 +1,7 @@
+
+[dev]
+pytest>=8.0
+pytest-asyncio>=0.23.0
+pytest-cov>=5.0
+ruff>=0.5
+mypy>=1.8

cogno_meter-0.1.0/cogno_meter.egg-info/top_level.txt

@@ -0,0 +1 @@
+cogno_meter

cogno_meter-0.1.0/pyproject.toml

@@ -0,0 +1,67 @@
+[project]
+name = "cogno-meter"
+version = "0.1.0"
+description = "Usage metering + pricing/overage for the Cogno cognitive pipeline — cross-modality (LLM/embedding tokens + STT/TTS chars) billable-token accounting against a plan allowance"
+readme = "README.md"
+requires-python = ">=3.10"
+license = "Apache-2.0"
+license-files = ["LICENSE"]
+authors = [{ name = "Vinicius Vale" }]
+maintainers = [{ name = "Sudoers AI" }]
+keywords = ["metering", "billing", "pricing", "tokens", "usage", "llm", "overage", "agents"]
+classifiers = [
+    "Development Status :: 3 - Alpha",
+    "Intended Audience :: Developers",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Topic :: Office/Business :: Financial",
+    "Topic :: Software Development :: Libraries",
+    "Operating System :: OS Independent",
+    "Typing :: Typed",
+]
+# Pure code: no third-party runtime dependency. The host loads its pricing config
+# (YAML/JSON/DB) and hands a plain mapping to PriceBook.from_mapping.
+dependencies = []
+
+[project.urls]
+Homepage = "https://github.com/sudoers-ai/cogno-meter"
+Repository = "https://github.com/sudoers-ai/cogno-meter"
+Issues = "https://github.com/sudoers-ai/cogno-meter/issues"
+
+[project.optional-dependencies]
+dev = [
+    "pytest>=8.0",
+    "pytest-asyncio>=0.23.0",
+    "pytest-cov>=5.0",
+    "ruff>=0.5",
+    "mypy>=1.8",
+]
+
+[build-system]
+requires = ["setuptools>=77.0.3", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[tool.setuptools.packages.find]
+include = ["cogno_meter*"]
+
+[tool.setuptools.package-data]
+cogno_meter = ["py.typed"]
+
+[tool.pytest.ini_options]
+asyncio_mode = "auto"
+testpaths = ["tests"]
+
+[tool.ruff]
+line-length = 110
+target-version = "py310"
+
+[tool.ruff.lint]
+select = ["E4", "E7", "E9", "F"]
+
+[tool.mypy]
+python_version = "3.10"
+ignore_missing_imports = true
+files = ["cogno_meter"]
+warn_unused_ignores = true

cogno_meter-0.1.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+