pickaxetax 0.1.0__tar.gz

pickaxetax-0.1.0/LICENSE

@@ -0,0 +1,202 @@
+
+                                 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 [yyyy] [name of copyright owner]
+
+   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.

pickaxetax-0.1.0/PKG-INFO

@@ -0,0 +1,179 @@
+Metadata-Version: 2.4
+Name: pickaxetax
+Version: 0.1.0
+Summary: Pickaxe Tax: measure and cut wasted AI compute -- local LLM proxy, conversation skeletons, over-computation benchmark, Compute Bubble Index.
+License-Expression: Apache-2.0
+Project-URL: Homepage, https://github.com/graviton94/pickaxetax
+Project-URL: Web app, https://graviton94.github.io/pickaxetax/
+Project-URL: Source, https://github.com/graviton94/pickaxetax
+Project-URL: Issues, https://github.com/graviton94/pickaxetax/issues
+Keywords: llm,tokens,efficiency,proxy,openai,anthropic,benchmark,sustainability
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Topic :: Internet :: Proxy Servers
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: fastapi>=0.110
+Requires-Dist: uvicorn>=0.27
+Requires-Dist: httpx>=0.27
+Requires-Dist: python-multipart>=0.0.9
+Provides-Extra: dev
+Requires-Dist: pytest>=8; extra == "dev"
+Provides-Extra: tiktoken
+Requires-Dist: tiktoken>=0.7; extra == "tiktoken"
+Dynamic: license-file
+
+# Pickaxe Tax
+
+> **Stop paying the pickaxe tax.** `#AntiTokenMaxing`
+> In a gold rush, the people selling pickaxes get rich. Your AI re-reads the whole conversation every single turn, and you pay for every re-read.
+> Build less. Waste less. Use the compute we already have.
+
+
+**Try it in your browser (nothing is uploaded): https://graviton94.github.io/pickaxetax/** · `pip install pickaxetax`
+
+[한국어](#한국어) · [Charter (KO)](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/CHARTER.md) · [Decisions](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/decisions/README.md) · [Architecture](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/ARCHITECTURE.md) · [Privacy](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/PRIVACY.md)
+
+**Chat models re-read the entire conversation on every turn.** Late constraints, re-asks, "thanks!" messages and topic switches inside one chat make that cost grow quadratically. Pickaxe Tax takes a shared conversation (share link, export JSON or pasted transcript) and **throws the text away**. What it keeps is a graph skeleton: agenda (topic labels), structure (per-turn intents and typed edges such as DEEPENS, PIVOTS, CORRECTS and RETRIES), depth (topic threads and drill-down level) and token flow. It then shows:
+
+- actual vs. optimal compute (one focused chat per topic, with the waste removed) and a one-shot lower bound
+- where the compute leaked: thank-you messages, discarded answers, re-sent off-topic history
+- a one-shot prompt skeleton built only from topic labels
+- habits to change, ranked by measured waste
+
+Contributed skeletons feed a public, **k-anonymous** topic graph (a topic is shown only once it appears in ≥ k conversations) plus aggregate waste statistics. The analyzer itself calls **no LLM**: a project about saving tokens shouldn't spend them.
+
+```bash
+pip install -e ".[dev]"
+pickaxetax serve                      # conversation skeleton analyzer (web); `pxt` is a short alias
+pickaxetax proxy                      # local proxy: OpenAI-compatible /v1, Anthropic base URL
+pickaxetax ledger                     # tokens spent, context re-sent, tokens avoided
+pickaxetax bench run --base-url http://localhost:11434 --model llama3.2   # donate benchmark compute
+pytest
+```
+
+**Local proxy.** Point your client at it and it forwards to the real provider with your own key, which it never stores. It measures every request (input, output, reasoning, re-sent context). By default it also drops earlier pure thank-you exchanges from the history and answers a bare "thanks" locally instead of re-reading the whole chat. "ok"/"yes" are never short-circuited, because in agent workflows they mean "go ahead". Requests with tools are never rewritten. Opt-in: `--cache` reuses responses to identical temperature-0 requests, stored on your disk.
+
+**Web app (local-first).** `site/` is a static page that runs the same engine in your browser. You can paste text, drop an export file, or use a share link through a bookmarklet that reads the page you already have open. Its Content-Security-Policy (`connect-src 'none'`) makes it impossible for the page to send your conversation anywhere. It is deployed to GitHub Pages; to run it locally: `cd site && python3 -m http.server`.
+
+**How to help at zero cost:** see [CONTRIBUTING.md](https://github.com/graviton94/pickaxetax/blob/HEAD/CONTRIBUTING.md).
+
+See [docs/ARCHITECTURE.md](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/ARCHITECTURE.md) and [docs/PRIVACY.md](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/PRIVACY.md).
+
+### Where this is going
+
+The AI boom is turning into a gold rush where only the pick-and-shovel sellers win: more GPUs, more data centers, more power. The waste is not only in conversations. It runs through the whole compute stack: infrastructure built ahead of revenue, oversized models and needless reasoning tokens, and wasteful usage. Pickaxe Tax is the open toolkit of **#AntiTokenMaxing**, a global effort to **prove that with public data and remove it with open tools**:
+
+- **Shared brain**: solution-path graph + open efficiency dataset, built from conversation *skeletons* contributed by users worldwide
+- **Prompt compiler**: a local proxy/SDK and coding-agent plugin that compile wasteful requests into minimal ones and measure the savings
+- **Watchdog — Compute Bubble Index**: how much of the deployed AI compute actually produced value? Top-down capital data (capex vs. AI revenue, circular deals, depreciation assumptions) meets bottom-up measurements (model over-computation, avoidable usage), published as a dashboard and a quarterly "gold rush bill" report ([design, KO](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/BUBBLE_INDEX.md))
+- **Standard**: a Token Efficiency Index for models and products
+
+Principles: local-first (raw text never leaves your device), two-step consent, no LLM calls in the analysis path, cold data and bold messages, and a mirror rather than an enemy to model developers. Public goods (data, standard, core engine) are held by a foundation under irrevocable open licenses. Full charter: [docs/CHARTER.md](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/CHARTER.md).
+
+## License
+
+Code: [Apache-2.0](https://github.com/graviton94/pickaxetax/blob/HEAD/LICENSE) · Data: [ODbL 1.0](https://github.com/graviton94/pickaxetax/blob/HEAD/DATA_LICENSE.md)
+
+---
+
+## 한국어
+
+**Pickaxe Tax (곡괭이세)**: 골드러시에서 돈을 버는 건 곡괭이 판매자다. 우리는 그 세금을 그만 낸다. 운동명은 **#AntiTokenMaxing**이다.
+
+> **AI는 같은 대화를 매 턴 처음부터 다시 읽습니다.**
+> 원문은 버리고, 대화의 *모양*만 남겨, 불필요했던 연산을 드러냅니다.
+
+프로젝트의 핵심 가치, 방향성, 조직 구조는 [프로젝트 헌장](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/CHARTER.md)에 정리되어 있습니다.
+
+### 왜 만드나
+
+"토큰맥싱"(token-maxing)은 습관입니다. 조건을 나중에 덧붙이고, 같은 질문을 다시 하고, 답이 끝난 뒤 "고마워"를 보내고, 한 창에서 주제를 바꿔 계속 이어 갑니다. 채팅형 AI는 답할 때마다 **이전 대화 전체를 다시 입력으로 읽기** 때문에 이런 습관의 비용은 대화 길이에 따라 제곱으로 커집니다. 40번째 턴의 "고마워" 한 마디는 수만 토큰을 다시 처리하게 만듭니다.
+
+이렇게 불어난 연산은 GPU, 데이터센터, 전력 수요로 이어집니다. 이 프로젝트는 그 낭비를 **측정 가능한 숫자**로 보여 주고, 다음 대화에서 바로 쓸 수 있는 습관과 프롬프트 골격으로 되돌려 주는 공익 프로젝트입니다. 배경과 원칙은 [docs/MANIFESTO.md](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/MANIFESTO.md)에 있습니다.
+
+### 무엇을 하나
+
+1. **올리기**: ChatGPT·Claude·Gemini 등의 공유 링크, 데이터 내보내기 JSON, 또는 `User:` / `Assistant:` 형식의 텍스트
+2. **골격 추출**: 원문을 메모리에서만 분석해 다음 항목을 뽑습니다
+   - **아젠다**: 대화의 핵심 주제 라벨
+   - **스트럭처**: 턴마다의 의도(질문·지시·보충·정정·재질문·계속·감사·대용량 붙여넣기)와 턴 사이의 관계(DEEPENS, PIVOTS, RETURNS, CORRECTS, RETRIES …)
+   - **뎁스**: 주제 갈래(branch)와 각 갈래 안의 파고들기 깊이
+   - **토큰 흐름**: 보이는 토큰, 실제 과금 입력(이력 재전송), 출력
+3. **원문 폐기**: 텍스트는 저장하지 않습니다. 저장되는 것은 골격(숫자, 의도 라벨, 주제 라벨, 솔트된 지문)뿐입니다
+4. **그래프 DB화**: `(:Conversation)-[:HAS_TURN]->(:Turn)-[:ABOUT]->(:Topic)` 형태의 속성 그래프로 저장합니다. Neo4j로 보낼 수 있는 Cypher 내보내기도 제공합니다
+5. **되돌려 주기**
+   - 실제 연산과 최적 연산 비교 (주제별 새 대화 + 낭비 제거 기준, 한 번에 물었을 때의 하한선)
+   - 낭비 유형별 비중: 감사 메시지, 버려진 답변, 무관한 주제 재전송
+   - 주제 라벨만으로 만든 **한 번에 끝내는 프롬프트 골격**
+   - 측정된 낭비 순으로 정렬한 습관 교정 팁
+6. **공익 그래프**: 기여된 골격을 집계해 주제 동시출현 그래프와 전체 낭비 통계를 공개합니다. **k-익명성**을 적용해 k개(기본 3개) 이상의 대화에서 등장한 주제만 노출합니다
+
+분석기 자체는 **LLM을 호출하지 않습니다.** 규칙 기반 의도 분류, 키워드 랭킹, SimHash를 쓰므로 분석에 드는 연산은 거의 0입니다.
+
+### 빠른 시작
+
+```bash
+pip install -e ".[dev]"
+pickaxetax serve                    # http://127.0.0.1:8000
+# 또는
+pickaxetax analyze chat.txt         # 로컬 분석 (기본값은 저장 안 함)
+pickaxetax analyze conversations.json --save
+pickaxetax analyze https://chatgpt.com/share/... --json
+pickaxetax export > public_graph.cypher
+pytest
+```
+
+| 환경 변수 | 기본값 | 설명 |
+|---|---|---|
+| `PICKAXETAX_DB` | `data/pickaxetax.sqlite3` | 그래프 저장소 경로 |
+| `PICKAXETAX_SECRET` | 자동 생성 후 DB에 보관 | ID와 지문에 쓰는 솔트. 운영 환경에서는 직접 지정 권장 |
+| `PICKAXETAX_K_ANON` | `3` | 공개 그래프의 k-익명성 임계값 (API로는 낮출 수 없음) |
+| `PICKAXETAX_INPUT_WEIGHT` | `0.25` | 입력 토큰 연산 가중치 (출력 = 1) |
+| `PICKAXETAX_WH_PER_1K` | `0.5` | 1k 연산 단위당 Wh. **예시용 가정치** |
+
+### API
+
+| 메서드 | 경로 | 설명 |
+|---|---|---|
+| POST | `/api/analyze` | `{"input": "<링크|JSON|텍스트>", "save": true}` |
+| POST | `/api/analyze/file` | multipart 파일 업로드 (`file`, `save`) |
+| GET | `/api/conversations/{id}` | 골격 조회 |
+| GET | `/api/conversations/{id}/graph.json` · `/cypher` | 내보내기 |
+| DELETE | `/api/conversations/{id}` | `X-Delete-Token` 헤더로 내 기여 삭제 |
+| GET | `/api/stats` | 전체 임팩트 집계 |
+| GET | `/api/topics` · `/api/topics/cypher` | k-익명 주제 그래프 |
+
+대화형 문서는 `/api/docs`에 있습니다.
+
+### 구조
+
+```
+pickaxetax/
+  ingest/     공유 링크(SSRF 방지 허용 목록), 내보내기 JSON, 텍스트 파서
+  analyze/    의도 분류 · 깊이/갈래 · 토큰 회계 · 제안 생성 (LLM 미사용)
+  graph/      SQLite 속성 그래프 저장소 + Cypher/JSON 내보내기
+  web/        FastAPI + 의존성 없는 단일 페이지 UI (한/영)
+  cli.py
+docs/         MANIFESTO · ARCHITECTURE · PRIVACY
+```
+
+자세한 설계는 [docs/ARCHITECTURE.md](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/ARCHITECTURE.md), 개인정보 설계는 [docs/PRIVACY.md](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/PRIVACY.md)를 보세요.
+
+### 로드맵
+
+- [ ] 브라우저 확장 또는 WASM으로 **클라이언트 측 골격 추출** (원문이 서버로 아예 가지 않게)
+- [ ] 공유 페이지 파서 강화 (클라이언트 렌더링되는 ChatGPT와 Gemini 신규 포맷)
+- [ ] 집계 통계에 차등 프라이버시 노이즈 추가
+- [ ] 주제별 "최단 경로" 패턴 라이브러리: 같은 아젠다를 가장 적은 턴으로 끝낸 구조 공유
+- [ ] 다국어 형태소 처리 (일본어, 중국어 분절)
+- [ ] Neo4j·Memgraph 백엔드 어댑터, 공개 데이터셋 정기 배포
+- [ ] 측정 기반 에너지 계수 (공개 벤치마크 연동)
+- [ ] 요청 속도 제한과 악용 방지
+
+### 라이선스
+
+코드는 [Apache-2.0](https://github.com/graviton94/pickaxetax/blob/HEAD/LICENSE), 데이터는 [ODbL 1.0](https://github.com/graviton94/pickaxetax/blob/HEAD/DATA_LICENSE.md)을 따릅니다.

pickaxetax-0.1.0/README.md

@@ -0,0 +1,151 @@
+# Pickaxe Tax
+
+> **Stop paying the pickaxe tax.** `#AntiTokenMaxing`
+> In a gold rush, the people selling pickaxes get rich. Your AI re-reads the whole conversation every single turn, and you pay for every re-read.
+> Build less. Waste less. Use the compute we already have.
+
+
+**Try it in your browser (nothing is uploaded): https://graviton94.github.io/pickaxetax/** · `pip install pickaxetax`
+
+[한국어](#한국어) · [Charter (KO)](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/CHARTER.md) · [Decisions](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/decisions/README.md) · [Architecture](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/ARCHITECTURE.md) · [Privacy](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/PRIVACY.md)
+
+**Chat models re-read the entire conversation on every turn.** Late constraints, re-asks, "thanks!" messages and topic switches inside one chat make that cost grow quadratically. Pickaxe Tax takes a shared conversation (share link, export JSON or pasted transcript) and **throws the text away**. What it keeps is a graph skeleton: agenda (topic labels), structure (per-turn intents and typed edges such as DEEPENS, PIVOTS, CORRECTS and RETRIES), depth (topic threads and drill-down level) and token flow. It then shows:
+
+- actual vs. optimal compute (one focused chat per topic, with the waste removed) and a one-shot lower bound
+- where the compute leaked: thank-you messages, discarded answers, re-sent off-topic history
+- a one-shot prompt skeleton built only from topic labels
+- habits to change, ranked by measured waste
+
+Contributed skeletons feed a public, **k-anonymous** topic graph (a topic is shown only once it appears in ≥ k conversations) plus aggregate waste statistics. The analyzer itself calls **no LLM**: a project about saving tokens shouldn't spend them.
+
+```bash
+pip install -e ".[dev]"
+pickaxetax serve                      # conversation skeleton analyzer (web); `pxt` is a short alias
+pickaxetax proxy                      # local proxy: OpenAI-compatible /v1, Anthropic base URL
+pickaxetax ledger                     # tokens spent, context re-sent, tokens avoided
+pickaxetax bench run --base-url http://localhost:11434 --model llama3.2   # donate benchmark compute
+pytest
+```
+
+**Local proxy.** Point your client at it and it forwards to the real provider with your own key, which it never stores. It measures every request (input, output, reasoning, re-sent context). By default it also drops earlier pure thank-you exchanges from the history and answers a bare "thanks" locally instead of re-reading the whole chat. "ok"/"yes" are never short-circuited, because in agent workflows they mean "go ahead". Requests with tools are never rewritten. Opt-in: `--cache` reuses responses to identical temperature-0 requests, stored on your disk.
+
+**Web app (local-first).** `site/` is a static page that runs the same engine in your browser. You can paste text, drop an export file, or use a share link through a bookmarklet that reads the page you already have open. Its Content-Security-Policy (`connect-src 'none'`) makes it impossible for the page to send your conversation anywhere. It is deployed to GitHub Pages; to run it locally: `cd site && python3 -m http.server`.
+
+**How to help at zero cost:** see [CONTRIBUTING.md](https://github.com/graviton94/pickaxetax/blob/HEAD/CONTRIBUTING.md).
+
+See [docs/ARCHITECTURE.md](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/ARCHITECTURE.md) and [docs/PRIVACY.md](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/PRIVACY.md).
+
+### Where this is going
+
+The AI boom is turning into a gold rush where only the pick-and-shovel sellers win: more GPUs, more data centers, more power. The waste is not only in conversations. It runs through the whole compute stack: infrastructure built ahead of revenue, oversized models and needless reasoning tokens, and wasteful usage. Pickaxe Tax is the open toolkit of **#AntiTokenMaxing**, a global effort to **prove that with public data and remove it with open tools**:
+
+- **Shared brain**: solution-path graph + open efficiency dataset, built from conversation *skeletons* contributed by users worldwide
+- **Prompt compiler**: a local proxy/SDK and coding-agent plugin that compile wasteful requests into minimal ones and measure the savings
+- **Watchdog — Compute Bubble Index**: how much of the deployed AI compute actually produced value? Top-down capital data (capex vs. AI revenue, circular deals, depreciation assumptions) meets bottom-up measurements (model over-computation, avoidable usage), published as a dashboard and a quarterly "gold rush bill" report ([design, KO](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/BUBBLE_INDEX.md))
+- **Standard**: a Token Efficiency Index for models and products
+
+Principles: local-first (raw text never leaves your device), two-step consent, no LLM calls in the analysis path, cold data and bold messages, and a mirror rather than an enemy to model developers. Public goods (data, standard, core engine) are held by a foundation under irrevocable open licenses. Full charter: [docs/CHARTER.md](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/CHARTER.md).
+
+## License
+
+Code: [Apache-2.0](https://github.com/graviton94/pickaxetax/blob/HEAD/LICENSE) · Data: [ODbL 1.0](https://github.com/graviton94/pickaxetax/blob/HEAD/DATA_LICENSE.md)
+
+---
+
+## 한국어
+
+**Pickaxe Tax (곡괭이세)**: 골드러시에서 돈을 버는 건 곡괭이 판매자다. 우리는 그 세금을 그만 낸다. 운동명은 **#AntiTokenMaxing**이다.
+
+> **AI는 같은 대화를 매 턴 처음부터 다시 읽습니다.**
+> 원문은 버리고, 대화의 *모양*만 남겨, 불필요했던 연산을 드러냅니다.
+
+프로젝트의 핵심 가치, 방향성, 조직 구조는 [프로젝트 헌장](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/CHARTER.md)에 정리되어 있습니다.
+
+### 왜 만드나
+
+"토큰맥싱"(token-maxing)은 습관입니다. 조건을 나중에 덧붙이고, 같은 질문을 다시 하고, 답이 끝난 뒤 "고마워"를 보내고, 한 창에서 주제를 바꿔 계속 이어 갑니다. 채팅형 AI는 답할 때마다 **이전 대화 전체를 다시 입력으로 읽기** 때문에 이런 습관의 비용은 대화 길이에 따라 제곱으로 커집니다. 40번째 턴의 "고마워" 한 마디는 수만 토큰을 다시 처리하게 만듭니다.
+
+이렇게 불어난 연산은 GPU, 데이터센터, 전력 수요로 이어집니다. 이 프로젝트는 그 낭비를 **측정 가능한 숫자**로 보여 주고, 다음 대화에서 바로 쓸 수 있는 습관과 프롬프트 골격으로 되돌려 주는 공익 프로젝트입니다. 배경과 원칙은 [docs/MANIFESTO.md](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/MANIFESTO.md)에 있습니다.
+
+### 무엇을 하나
+
+1. **올리기**: ChatGPT·Claude·Gemini 등의 공유 링크, 데이터 내보내기 JSON, 또는 `User:` / `Assistant:` 형식의 텍스트
+2. **골격 추출**: 원문을 메모리에서만 분석해 다음 항목을 뽑습니다
+   - **아젠다**: 대화의 핵심 주제 라벨
+   - **스트럭처**: 턴마다의 의도(질문·지시·보충·정정·재질문·계속·감사·대용량 붙여넣기)와 턴 사이의 관계(DEEPENS, PIVOTS, RETURNS, CORRECTS, RETRIES …)
+   - **뎁스**: 주제 갈래(branch)와 각 갈래 안의 파고들기 깊이
+   - **토큰 흐름**: 보이는 토큰, 실제 과금 입력(이력 재전송), 출력
+3. **원문 폐기**: 텍스트는 저장하지 않습니다. 저장되는 것은 골격(숫자, 의도 라벨, 주제 라벨, 솔트된 지문)뿐입니다
+4. **그래프 DB화**: `(:Conversation)-[:HAS_TURN]->(:Turn)-[:ABOUT]->(:Topic)` 형태의 속성 그래프로 저장합니다. Neo4j로 보낼 수 있는 Cypher 내보내기도 제공합니다
+5. **되돌려 주기**
+   - 실제 연산과 최적 연산 비교 (주제별 새 대화 + 낭비 제거 기준, 한 번에 물었을 때의 하한선)
+   - 낭비 유형별 비중: 감사 메시지, 버려진 답변, 무관한 주제 재전송
+   - 주제 라벨만으로 만든 **한 번에 끝내는 프롬프트 골격**
+   - 측정된 낭비 순으로 정렬한 습관 교정 팁
+6. **공익 그래프**: 기여된 골격을 집계해 주제 동시출현 그래프와 전체 낭비 통계를 공개합니다. **k-익명성**을 적용해 k개(기본 3개) 이상의 대화에서 등장한 주제만 노출합니다
+
+분석기 자체는 **LLM을 호출하지 않습니다.** 규칙 기반 의도 분류, 키워드 랭킹, SimHash를 쓰므로 분석에 드는 연산은 거의 0입니다.
+
+### 빠른 시작
+
+```bash
+pip install -e ".[dev]"
+pickaxetax serve                    # http://127.0.0.1:8000
+# 또는
+pickaxetax analyze chat.txt         # 로컬 분석 (기본값은 저장 안 함)
+pickaxetax analyze conversations.json --save
+pickaxetax analyze https://chatgpt.com/share/... --json
+pickaxetax export > public_graph.cypher
+pytest
+```
+
+| 환경 변수 | 기본값 | 설명 |
+|---|---|---|
+| `PICKAXETAX_DB` | `data/pickaxetax.sqlite3` | 그래프 저장소 경로 |
+| `PICKAXETAX_SECRET` | 자동 생성 후 DB에 보관 | ID와 지문에 쓰는 솔트. 운영 환경에서는 직접 지정 권장 |
+| `PICKAXETAX_K_ANON` | `3` | 공개 그래프의 k-익명성 임계값 (API로는 낮출 수 없음) |
+| `PICKAXETAX_INPUT_WEIGHT` | `0.25` | 입력 토큰 연산 가중치 (출력 = 1) |
+| `PICKAXETAX_WH_PER_1K` | `0.5` | 1k 연산 단위당 Wh. **예시용 가정치** |
+
+### API
+
+| 메서드 | 경로 | 설명 |
+|---|---|---|
+| POST | `/api/analyze` | `{"input": "<링크|JSON|텍스트>", "save": true}` |
+| POST | `/api/analyze/file` | multipart 파일 업로드 (`file`, `save`) |
+| GET | `/api/conversations/{id}` | 골격 조회 |
+| GET | `/api/conversations/{id}/graph.json` · `/cypher` | 내보내기 |
+| DELETE | `/api/conversations/{id}` | `X-Delete-Token` 헤더로 내 기여 삭제 |
+| GET | `/api/stats` | 전체 임팩트 집계 |
+| GET | `/api/topics` · `/api/topics/cypher` | k-익명 주제 그래프 |
+
+대화형 문서는 `/api/docs`에 있습니다.
+
+### 구조
+
+```
+pickaxetax/
+  ingest/     공유 링크(SSRF 방지 허용 목록), 내보내기 JSON, 텍스트 파서
+  analyze/    의도 분류 · 깊이/갈래 · 토큰 회계 · 제안 생성 (LLM 미사용)
+  graph/      SQLite 속성 그래프 저장소 + Cypher/JSON 내보내기
+  web/        FastAPI + 의존성 없는 단일 페이지 UI (한/영)
+  cli.py
+docs/         MANIFESTO · ARCHITECTURE · PRIVACY
+```
+
+자세한 설계는 [docs/ARCHITECTURE.md](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/ARCHITECTURE.md), 개인정보 설계는 [docs/PRIVACY.md](https://github.com/graviton94/pickaxetax/blob/HEAD/docs/PRIVACY.md)를 보세요.
+
+### 로드맵
+
+- [ ] 브라우저 확장 또는 WASM으로 **클라이언트 측 골격 추출** (원문이 서버로 아예 가지 않게)
+- [ ] 공유 페이지 파서 강화 (클라이언트 렌더링되는 ChatGPT와 Gemini 신규 포맷)
+- [ ] 집계 통계에 차등 프라이버시 노이즈 추가
+- [ ] 주제별 "최단 경로" 패턴 라이브러리: 같은 아젠다를 가장 적은 턴으로 끝낸 구조 공유
+- [ ] 다국어 형태소 처리 (일본어, 중국어 분절)
+- [ ] Neo4j·Memgraph 백엔드 어댑터, 공개 데이터셋 정기 배포
+- [ ] 측정 기반 에너지 계수 (공개 벤치마크 연동)
+- [ ] 요청 속도 제한과 악용 방지
+
+### 라이선스
+
+코드는 [Apache-2.0](https://github.com/graviton94/pickaxetax/blob/HEAD/LICENSE), 데이터는 [ODbL 1.0](https://github.com/graviton94/pickaxetax/blob/HEAD/DATA_LICENSE.md)을 따릅니다.

pickaxetax-0.1.0/pickaxetax/__init__.py

@@ -0,0 +1,7 @@
+"""Pickaxe Tax: measure and cut wasted AI compute. #AntiTokenMaxing"""
+
+from .analyze import analyze
+from .ingest import ingest_text
+
+__version__ = "0.1.0"
+__all__ = ["analyze", "ingest_text", "__version__"]

pickaxetax-0.1.0/pickaxetax/analyze/__init__.py

@@ -0,0 +1,90 @@
+"""RawConversation -> Skeleton. The raw text does not leave this function."""
+
+from __future__ import annotations
+
+import hashlib
+import hmac
+from collections import Counter
+from datetime import datetime, timezone
+
+from ..models import ASSISTANT, USER, RawConversation, Skeleton, TurnNode
+from ..text import detect_language, jaccard, rank_keywords, simhash, simhash_similarity, split_code, words
+from ..tokens import estimate_tokens
+from . import intent as I
+from .depth import build_structure
+from .suggest import suggest
+from .waste import account
+
+AGENDA_SIZE = 7
+USER_TOPICS = 5
+ASSISTANT_TOPICS = 3
+
+
+def analyze(raw: RawConversation, salt: bytes = b"") -> Skeleton:
+    turns_raw = [t for t in raw.turns if t.role in (USER, ASSISTANT) and t.text.strip()]
+    if not any(t.role == USER for t in turns_raw):
+        raise ValueError("conversation has no user turns")
+
+    language = detect_language(" ".join(t.text[:2000] for t in turns_raw))
+    ranked = rank_keywords(
+        [t.text for t in turns_raw if t.role == USER],
+        [t.text for t in turns_raw if t.role == ASSISTANT],
+    )
+    score = dict(ranked)
+    agenda = [w for w, _ in ranked[:AGENDA_SIZE]]
+
+    nodes: list[TurnNode] = []
+    user_prints: list[tuple[int, int, set]] = []  # (index, simhash, word set)
+    retry_of: dict[int, int] = {}
+    cues: dict[int, set[str]] = {}  # transient, never stored
+
+    for idx, rt in enumerate(turns_raw):
+        tokens = estimate_tokens(rt.text)
+        _, has_code = split_code(rt.text)
+        ws = words(rt.text)
+        wset = set(ws)
+        limit = USER_TOPICS if rt.role == USER else ASSISTANT_TOPICS
+        topics = sorted((w for w in wset if w in score), key=lambda w: (-score[w], w))[:limit]
+        fp = simhash(rt.text, key=salt)
+        cues[idx] = wset if rt.role == USER else {w for w, _ in Counter(ws).most_common(8)}
+
+        if rt.role == USER:
+            first = not user_prints
+            similar = False
+            if len(wset) >= 3:
+                for j, jfp, jset in reversed(user_prints):
+                    if simhash_similarity(fp, jfp) >= 0.9 or jaccard(wset, jset) >= 0.8:
+                        retry_of[idx] = j
+                        similar = True
+                        break
+            user_prints.append((idx, fp, wset))
+            kind = I.classify_user(rt.text, tokens, similar, has_code, first=first)
+        else:
+            kind = I.classify_assistant(rt.text, tokens, has_code)
+
+        nodes.append(
+            TurnNode(index=idx, role=rt.role, tokens=tokens, intent=kind, topics=topics,
+                     has_code=has_code, fingerprint=f"{fp:016x}")
+        )
+
+    edges = build_structure(nodes, retry_of, cues)
+    cues.clear()
+    metrics = account(nodes)
+
+    digest = hmac.new(salt, "|".join(f"{n.role}:{n.fingerprint}:{n.tokens}" for n in nodes).encode(),
+                      hashlib.blake2b).hexdigest()[:20]
+    sk = Skeleton(
+        id=digest,
+        source=raw.source,
+        language=language,
+        agenda=agenda,
+        turns=nodes,
+        edges=edges,
+        metrics=metrics,
+        created_at=datetime.now(timezone.utc).replace(microsecond=0).isoformat(),
+    )
+    sk.suggestion = suggest(sk)
+    return sk
+
+
+__all__ = ["analyze"]