zing-audit 0.2.0__tar.gz

zing_audit-0.2.0/.gitignore

@@ -0,0 +1,32 @@
+# Python
+__pycache__/
+*.py[cod]
+*.egg-info/
+.eggs/
+build/
+dist/
+*.whl
+
+# Virtual envs
+.venv/
+venv/
+env/
+
+# Tooling caches
+.pytest_cache/
+.ruff_cache/
+.mypy_cache/
+.coverage
+htmlcov/
+
+# zing output
+reports/
+*.zing.json
+
+# Reference material (ChatGPT v0 seed, not part of the project)
+tmp/
+
+# OS / editor
+.DS_Store
+.idea/
+.vscode/

zing_audit-0.2.0/CHANGELOG.md

@@ -0,0 +1,64 @@
+# Changelog
+
+All notable changes to **zing** are documented here. The format is based on
+[Keep a Changelog](https://keepachangelog.com/en/1.1.0/), and this project adheres
+to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [Unreleased]
+
+## [0.2.0] — Anthropic support & roadmap security detectors
+
+### Added
+
+- **Anthropic Messages API support.** zing now audits Anthropic-native relays
+  (`/v1/messages`) as well as OpenAI Chat Completions, behind one detector
+  interface. The protocol is auto-detected from the base_url/model or forced with
+  `--api openai|anthropic` (and `--target-api` / `--baseline-api` for `compare`).
+- **Three new `deep`-suite security detectors** (formerly roadmap):
+  - `injected_prompt` — detects a hidden, silently-prepended system prompt from a
+    large *fixed* input-token overhead (measured across two message sizes) plus a
+    leak probe; needs both signals to warn.
+  - `integrity` — known-answer URL/package canaries catch in-flight response/tool-call
+    tampering (value substituted, structure preserved). CRITICAL only when a trusted
+    baseline returns the canary intact; otherwise MEDIUM.
+  - `prompt_cache` — flags prompt-prefix caching by TTFT timing (informational; states
+    that cross-user cache sharing is not provable from a single key).
+- Automated PyPI publishing via GitHub Actions + Trusted Publishing
+  (`.github/workflows/release.yml`); see `docs/PUBLISHING.md`.
+
+## [0.1.0] — first public alpha
+
+First public release. A local-first CLI that audits whether an OpenAI-compatible
+API relay (中转站 / reseller / proxy) actually serves the model it claims to —
+or quietly substitutes a cheaper one, truncates the context window, fakes
+streaming, or inflates token billing (货不对板检测).
+
+### Added
+
+- `zing check` — audit one relay endpoint against what it claims (model id +
+  optional provider hint).
+- `zing compare` — audit a relay against a trusted baseline of the same declared
+  model (the strongest downgrade evidence).
+- `zing models` — probe an endpoint's `GET /v1/models` list.
+- `zing kb` — inspect the bundled knowledge base.
+- `zing init` — write a starter `zing.yaml` config.
+- Eleven detectors across nine scored dimensions: model identity & downgrade
+  fingerprinting, real context window & truncation, capability claims, token/usage
+  billing, streaming authenticity, OpenAI-protocol conformance, determinism/cache
+  correctness, concurrent reliability, transport/secret-handling security, plus
+  connectivity and an optional LLM-judged quality assessment (`--judge`).
+- Bundled knowledge base of **85 native model profiles across 7 platforms**
+  (OpenAI, Anthropic, Google Gemini, DeepSeek, Qwen, GLM, Moonshot), editable as
+  YAML and overridable via `--kb-dir` / `ZING_KB_DIR`.
+- Two detection modes: pure-code deterministic probes (default) and a code+LLM
+  hybrid that consults a separate trusted judge model.
+- JSON, Markdown, and HTML reports; `--json` for machine/agent consumption.
+- Evidence-first verdicts (CLEAN / LOW / MEDIUM / HIGH / INCONCLUSIVE) with
+  confidence, designed to avoid false accusations of honest relays.
+- Secret hygiene: API keys are fingerprinted (never stored) and relay-controlled
+  text is redacted before it reaches any report (JSON/Markdown/HTML).
+- `--fail-under` / `--fail-on-risk` exit-code gates for CI use.
+
+[Unreleased]: https://github.com/cenbonew/zing/compare/v0.2.0...HEAD
+[0.2.0]: https://github.com/cenbonew/zing/compare/v0.1.0...v0.2.0
+[0.1.0]: https://github.com/cenbonew/zing/releases/tag/v0.1.0

zing_audit-0.2.0/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 zing contributors
+
+   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.

zing_audit-0.2.0/NOTICE

@@ -0,0 +1,10 @@
+zing — LLM relay reality check
+Copyright 2026 zing contributors
+
+This product is licensed under the Apache License, Version 2.0 (see the LICENSE
+file). You may not use this product except in compliance with that License.
+
+zing is a black-box auditing aid. It gathers reproducible evidence of behavioral
+divergence and risk — not cryptographic proof of model identity or fraud. See
+README.md ("Responsible use") and docs/METHODOLOGY.md before publishing any
+report that names a vendor.

zing_audit-0.2.0/PKG-INFO

@@ -0,0 +1,230 @@
+Metadata-Version: 2.4
+Name: zing-audit
+Version: 0.2.0
+Summary: LLM relay reality check — a CLI that audits whether an API relay actually serves the model it claims (货不对板检测).
+Project-URL: Homepage, https://github.com/cenbonew/zing
+Project-URL: Repository, https://github.com/cenbonew/zing
+Project-URL: Issues, https://github.com/cenbonew/zing/issues
+Author: zing contributors
+License-Expression: Apache-2.0
+License-File: LICENSE
+License-File: NOTICE
+Keywords: api,audit,context-window,llm,model-fingerprinting,openai,proxy,relay
+Classifier: Development Status :: 3 - Alpha
+Classifier: Environment :: Console
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Software Development :: Quality Assurance
+Requires-Python: >=3.10
+Requires-Dist: httpx>=0.27
+Requires-Dist: pydantic>=2.6
+Requires-Dist: pyyaml>=6.0
+Requires-Dist: rich>=13.7
+Requires-Dist: typer>=0.12
+Provides-Extra: dev
+Requires-Dist: mypy>=1.10; extra == 'dev'
+Requires-Dist: pytest-asyncio>=0.23; extra == 'dev'
+Requires-Dist: pytest>=8.0; extra == 'dev'
+Requires-Dist: ruff>=0.5; extra == 'dev'
+Requires-Dist: types-pyyaml>=6.0; extra == 'dev'
+Provides-Extra: tokenizers
+Requires-Dist: tiktoken>=0.7; extra == 'tokenizers'
+Description-Content-Type: text/markdown
+
+# zing — LLM relay reality check
+
+> **English** · [中文](README.zh-CN.md)
+
+[![CI](https://github.com/cenbonew/zing/actions/workflows/ci.yml/badge.svg)](https://github.com/cenbonew/zing/actions/workflows/ci.yml)
+[![License: Apache-2.0](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](LICENSE)
+[![Python 3.10+](https://img.shields.io/badge/python-3.10%2B-blue.svg)](pyproject.toml)
+
+**zing** is a local-first CLI that audits whether an API relay (中转站 / reseller /
+proxy) actually serves the model it claims to — or quietly substitutes a cheaper
+one, truncates your context window, fakes streaming, or inflates token billing
+(**货不对板检测**). It speaks both **OpenAI Chat Completions** and the
+**Anthropic Messages API** (auto-detected, or forced with `--api`).
+
+You point it at a relay endpoint and the model it advertises; zing runs a battery
+of black-box probes, compares the observed behavior against a bundled knowledge
+base of **85 native model profiles across 7 platforms**, and prints a clear,
+evidence-backed verdict — for a human, or as JSON for another tool / LLM to read.
+
+> zing reports **black-box evidence of divergence and risk, not cryptographic
+> proof of fraud.** See [Responsible use](#responsible-use).
+
+---
+
+## Why
+
+The relay-key market is full of "GPT-4o for 1/10th the price" offers. Many are
+honest. Some are not — and the dishonest ones are hard to catch by eye:
+
+- You ask for `gpt-4o`; you're quietly served `gpt-4o-mini` or an open model.
+- The relay advertises a 1M-token context but silently truncates to 32K.
+- "Streaming" is the full response buffered and re-chunked, with no latency win.
+- Reported `usage` tokens are inflated, so your balance burns faster than it should.
+- A model that should support tool-calling / JSON mode quietly doesn't.
+
+zing turns "this feels off" into a reproducible report.
+
+## Install
+
+Requires Python 3.10+.
+
+```bash
+# from PyPI
+pip install zing-audit          # the `zing` command
+
+# or from source
+git clone https://github.com/cenbonew/zing
+cd zing
+pip install -e .
+```
+
+(Maintainers: see [docs/PUBLISHING.md](docs/PUBLISHING.md) for the release process.)
+
+Optional: install the `tokenizers` extra for accurate OpenAI-family token
+counting in the billing audit:
+
+```bash
+pip install -e '.[tokenizers]'
+```
+
+## Quick start
+
+```bash
+# 1) audit a relay against what it claims (model id + provider hint)
+export ZING_API_KEY=sk-your-relay-key
+zing check \
+  --base-url https://relay.example.com/v1 \
+  --api-key env:ZING_API_KEY \
+  --model gpt-4o \
+  --suite standard
+
+# 2) the strongest check: compare against a trusted baseline of the same model
+export OPENAI_API_KEY=sk-your-openai-key
+zing compare \
+  --target-base-url https://relay.example.com/v1 --target-api-key env:ZING_API_KEY --target-model gpt-4o \
+  --baseline-base-url https://api.openai.com/v1 --baseline-api-key env:OPENAI_API_KEY --baseline-model gpt-4o \
+  --suite deep
+
+# 3) audit an Anthropic-native (Messages API) relay — protocol is auto-detected
+#    from the base_url/model, or force it with --api anthropic
+zing check --base-url https://relay.example.com/v1 --model claude-opus-4-8 \
+  --api-key env:ZING_API_KEY --api anthropic
+
+# 4) inspect the bundled knowledge base
+zing kb            # all 85 models
+zing kb deepseek   # one provider
+
+# 5) generate a config you can commit
+zing init          # writes zing.yaml
+zing check -c zing.yaml
+```
+
+### As a tool for an LLM / agent
+
+`--json` prints the structured report to stdout instead of writing files — feed
+it straight to another program or model:
+
+```bash
+zing check --base-url ... --model gpt-4o --json | jq .verdict
+```
+
+## What it checks
+
+zing scores nine dimensions. The three that most directly reveal 货不对板
+(model identity, real context window, capability claims) carry the most weight.
+
+| Dimension | What it catches |
+|---|---|
+| **model_identity** | Silent model downgrade/substitution — self-identification, knowledge-cutoff, tokenizer fingerprints, the echoed `model` field |
+| **context_window** | Silent context truncation (claim 1M, recall fails at 32K) and lost-in-the-middle from cheap RAG/summarization shims, via needle-in-a-haystack + binary search |
+| **capability** | Tool-calling / JSON-mode / json-schema / max-output claims that aren't actually delivered (or *over*-delivered, hinting at a substitute) |
+| **billing** | Token/usage inflation and missing/unverifiable usage accounting, via an independent tokenizer estimate |
+| **streaming** | Fake streaming (buffer-then-chunk) detected from chunk count and inter-chunk timing |
+| **protocol** | OpenAI-compatibility conformance: multi-turn, stop sequences, response shape, error schema — and a determinism sub-check for response caching that ignores temperature/seed |
+| **reliability** | Concurrent success rate and latency (HTTP 429 throttling bucketed separately) |
+| **connectivity** | Endpoint reachability and the advertised `/v1/models` list |
+| **security** | Transport (HTTPS), header hygiene, secret echo; hidden injected system prompt (fixed input-token overhead + leak), in-flight response/tool-call tampering via known-answer canaries (URL/package substitution), and prompt-prefix caching (timing) |
+
+See [docs/METHODOLOGY.md](docs/METHODOLOGY.md) for the technique behind each check,
+which relay trick it maps to, and its false-positive caveats.
+
+## Two detection modes
+
+- **Pure code (default):** every deterministic probe — fingerprints, context
+  sweep, billing math, streaming timing. No second model needed; fully
+  reproducible.
+- **Code + LLM hybrid (`--judge`):** additionally consults a *trusted* judge
+  model (configured separately, never the target) to assess fuzzy signals like
+  quality and reasoning depth that pure code can't decide. Powers the
+  `quality_judge` detector.
+
+```bash
+zing check --base-url ... --model gpt-4o --suite deep --judge \
+  --judge-base-url https://api.openai.com/v1 --judge-api-key env:OPENAI_API_KEY --judge-model gpt-4o-mini
+```
+
+## Suites
+
+| Suite | Detectors | Cost |
+|---|---|---|
+| `smoke` | connectivity, security | very low |
+| `standard` | + protocol, model_identity, capability, streaming, billing, reliability | low–medium |
+| `deep` | + context_window, determinism, injected_prompt, integrity, prompt_cache, quality_judge (if `--judge`) | higher (long-context & timing probes cost tokens) |
+| `full` | everything | highest |
+
+The context-window probe is bounded by `--max-context-tokens` (default 200K) so
+auditing a 1M-token model stays affordable.
+
+## Example verdict
+
+```text
+╭─ ✗ HIGH RISK — Strong evidence the relay does not deliver the claimed model… ─╮
+│ Target : my-relay · model gpt-4o · provider openai                            │
+│ Mode   : check · suite deep                                                   │
+│ Score  : 53.5/100 (rating F) · confidence medium                             │
+│                                                                               │
+│ Overall health score 53.5/100. Findings: 3 high. …                            │
+╰───────────────────────────────────────────────────────────────────────────────╯
+  • Self-identifies as a rival brand (anthropic) under the claimed model id gpt-4o
+  • Real context window ~8000 << declared 128000 (silent truncation suspected)
+  • Reported prompt tokens far exceed independent estimate
+```
+
+Reports are written to `reports/` as JSON, Markdown, and HTML.
+
+## Knowledge base
+
+Profiles live in [`zing/knowledge/data/`](zing/knowledge/data) as editable YAML —
+one per provider (OpenAI, Anthropic, Google Gemini, DeepSeek, Qwen, GLM,
+Moonshot). Each model carries its native context window, max output, tokenizer,
+capability flags, identity keywords, and behavioral fingerprints. Add or override
+profiles without forking:
+
+```bash
+zing check --kb-dir ./my-profiles ...     # or set ZING_KB_DIR
+```
+
+## Responsible use
+
+zing is a black-box auditing aid. It **cannot prove**:
+
+- that a provider stores or trains on your prompts,
+- that it always routes to one exact model (relays can route probabilistically),
+- billing fraud beyond what independent token estimation can suggest.
+
+Use reports for your own due diligence. **Do not publicly accuse a vendor** based
+on a single run without reviewing sample size, cost settings, and local law. Run
+`zing compare` against a trusted baseline before drawing strong conclusions.
+
+## License
+
+[Apache-2.0](LICENSE)