Metadata-Version: 2.4
Name: callus
Version: 0.3.1
Summary: Per-author voice calibration: score AI tells and rewrite drafts to your natural voice.
Project-URL: Homepage, https://github.com/VDP89/callus
Project-URL: Issues, https://github.com/VDP89/callus/issues
Project-URL: Repository, https://github.com/VDP89/callus
Author-email: Victor Daniel Del Puerto Gauto <info@dgingenieriasrl.com>
License: 
                                         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.
License-File: LICENSE
Keywords: ai-detection,ai-writing,anti-ai-slop,claude-code,llm-tooling,voice-calibration
Classifier: Development Status :: 3 - Alpha
Classifier: Environment :: Console
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: End Users/Desktop
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
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: Topic :: Software Development
Classifier: Topic :: Text Processing :: Linguistic
Requires-Python: >=3.10
Requires-Dist: typer>=0.12
Provides-Extra: dev
Requires-Dist: pytest-cov>=4.1; extra == 'dev'
Requires-Dist: pytest>=8.0; extra == 'dev'
Requires-Dist: ruff>=0.4; extra == 'dev'
Description-Content-Type: text/markdown

<!-- markdownlint-disable MD033 -->
<div align="center">
  <img src="assets/callus-launch.gif" alt="callus" width="600" style="max-width: 100%; height: auto;" />
  <p><strong>Per-author voice calibration. Score AI tells and rewrite drafts toward your natural voice.</strong></p>

  <p>
    <a href="https://github.com/VDP89/callus/actions/workflows/ci.yml">
      <img alt="CI" src="https://github.com/VDP89/callus/actions/workflows/ci.yml/badge.svg">
    </a>
    <a href="LICENSE">
      <img alt="License" src="https://img.shields.io/badge/License-Apache%202.0-blue.svg">
    </a>
    <a href="https://claude.com/claude-code">
      <img alt="Claude Code" src="https://img.shields.io/badge/Claude%20Code-Compatible-6366f1">
    </a>
  </p>
</div>

> Status: **alpha (v0.3.0)**. Per-author voice profile, score, iterative rewriter, and packaged incremental-capture hook are working. Install with `pip install callus`, or use as a Python library / the `callus` CLI.

---

## Why this exists

I write in English as a second language. I tested four commercial AI detectors on my published blog posts. Each of them returned over 90% AI on prose I had written, edited, and corrected myself. A peer-reviewed paper from Stanford in 2023 explained why: detectors based on perplexity flag non-native English writers at 61% false positive, because the same features that mark "AI-like text" — limited vocabulary, common collocations — also describe how most non-native speakers write.

So the score number was useless to me as a metric for iteration. If I rewrote a paragraph to remove the actual AI tells — the aphorisms, the hinge phrases, the triplet negations — the score barely moved. The detector was measuring my passport, not my prose.

`callus` does the other thing. It compares your draft against *your own* raw writing — extracted from your Claude Code sessions or any other source where you typed unedited — and against a small library of AI tells. The score is "how far is this draft from your voice + how dense are the tells", not "what is the probability this came from an LLM". That distinction is the whole point.

---

## What it does

Three operations, all calibrated against you, not against a generic native-English baseline:

- **`callus score <file>`** — Returns a 0-100 score with a per-axis breakdown (voice_distance, tells_density, structural_ai_patterns), the language detected (EN/ES), and concrete tells cited verbatim from the draft with suggested fixes.
- **`callus rewrite <file> --target 25`** — Iteratively rewrites the draft using your voice corpus as few-shot context. Stops when it hits the target or starts degrading (early-stop on degrade). Preserves claims, numbers, and links; allows paragraph restructuring. Typical run: 2-3 iterations, $0.01-0.02 USD on Haiku.
- **`callus build-corpus --source <dir>`** — Extracts your raw user-typed prompts from Claude Code session logs and applies thirteen calibrated filters (drops pastes, command dumps, Codex reviews, dashboard copy, emoji-heavy reviewer output) so what ends up in the corpus is actually your voice, not your assistant's.

A fourth piece, **`callus approve <pending.md>`**, merges new candidates from incremental capture (see hook setup) after you mark each one OK / NO / MEH.

**`callus similarity <file>`** is a free, offline check: it reports how lexically close a draft is to your corpus (character n-gram cosine against your most similar samples), as a cheap complement to `callus score` when you do not want to spend an LLM call. Low similarity is a red flag worth a real look; high similarity does not by itself prove voice.

---

## Quick start

```bash
pip install callus
callus --version

# Extract your raw voice from Claude Code sessions
callus build-corpus --source ~/.claude/projects/your-project

# Score a draft
callus score path/to/draft.md

# Rewrite a draft toward your voice
callus rewrite path/to/draft.md --target 25 --out path/to/draft.rewritten.md
```

You need a working `claude` CLI on your PATH (the package shells out to `claude -p --model haiku` for scoring and rewriting).

---

## Why not just use GPTZero / Originality / Humalingo?

I ran the same blog post through Humalingo. It scored 91% AI. I ran the LessWrong submission of the same content, which has measurably more AI tells (hinge phrases, triplet negations, defensive clarifications), through Humalingo as well. It scored 92%.

A one-point delta between a draft with four BLOCK-severity tells and a draft with one. The classifier cannot see the difference between "clean voice" and "voice plus tells" within the cluster of AI-assisted writing. The custom judge in `callus` scored the same two drafts at 20 and 33 — a thirteen-point delta that maps onto what a human moderator would actually read for.

The Stanford 2023 result on non-native English bias ([arXiv:2304.02819](https://arxiv.org/abs/2304.02819)) explains why. Classifier-based detectors lean on perplexity, which is an artifact of vocabulary and collocation distribution. Native English essayists and AI both share a higher-perplexity distribution. Non-native writers and AI both share a lower-perplexity distribution. The detector cannot tell them apart structurally.

`callus` does not try to. It measures something else: distance from a specific writer's voice, defined by that writer's own raw text. There is no claim of universality; there is a claim of usefulness to the operator.

If you want to score against the generic native-English baseline, use Humalingo. If you want to iterate on a draft so it reads more like the way you actually write, use this.

More detail in [docs/why_not_classifier.md](docs/why_not_classifier.md).

---

## Setting up your voice

`callus` ships without a corpus on purpose. The whole architecture only works if the corpus is yours.

1. **Build the corpus** from your Claude Code sessions: `callus build-corpus --source <path>`.
2. **Sample-review** the first 16 entries by hand. The filters drop most contamination but you should know what is in your corpus.
3. **Write a voice profile** by copying [`examples/profile_template.md`](examples/profile_template.md) and editing the rules to match how you actually write (see [`examples/voz_victor.md`](examples/voz_victor.md) for a filled-in example). Point callus at it with `CALLUS_PROFILE=path/to/profile.md` or `--profile path/to/profile.md`. Until you do, callus uses a generic default profile — no author-specific rules — and tells you so on each run.
4. **Score and iterate**: `callus score draft.md --profile my_voice.md`.

Optionally set `CALLUS_AUTHOR="Your Name"` to label the judge prompt.

Full walkthrough: [docs/setup_your_voice.md](docs/setup_your_voice.md).

---

## Incremental capture (optional)

If you want the corpus to grow automatically every time you close a session in Claude Code, wire a hook:

```json
"UserPromptSubmit": [
  {
    "hooks": [
      {
        "type": "command",
        "command": "python -m callus.hooks.voice_corpus_close"
      }
    ]
  }
]
```

The hook (`callus.hooks.voice_corpus_close`, shipped in the package) watches for closing phrases ("cerramos", "guardar memoria", "wrap up", "save memory", ...) in your prompts. When it sees one, it reads the session transcript, applies the same thirteen filters as `build-corpus`, deduplicates against your existing corpus, and writes a pending review file. Nothing gets merged without you running `callus approve`. Configure with `CALLUS_PENDING_DIR` (where reviews land), `CALLUS_CLOSE_KEYWORDS` (extra triggers), and `CALLUS_CORPUS` (corpus to dedup against). It never blocks your session.

---

## How it is built

```
                ┌──────────────────────────────────────┐
                │              callus.score             │
                │   LLM-as-judge, multi-axis, EN+ES     │
                └───────────────┬──────────────────────┘
                                │
        ┌───────────────────────┴─────────────────────────┐
        │                                                  │
┌───────▼────────┐                              ┌─────────▼──────────┐
│  callus.rewrite │  ←── few-shot voice ──→     │  callus.build_corpus│
│   iterative     │       voice_corpus.jsonl     │   F1-F13 filters    │
│   loop          │                              │   (calibrated)       │
└────────────────┘                              └─────────────────────┘
```

The judge prompt sees four things on every call: your voice profile, a generic tells_ai library, six rotating raw-voice samples from your corpus, and the draft. It returns strict JSON with axis scores and verbatim citations. The rewriter feeds those citations back into a follow-up call that asks the LLM to produce a voice-translated draft while preserving every quantitative claim and link.

The bias correction for non-native EN is built into the prompt instructions, not as a post-hoc adjustment.

---

## When NOT to use callus

- You do not have a corpus of your own writing. The skill is calibration against you; without a corpus, you are scoring against nothing.
- You want a generic "is this AI" detector for a third party's writing. Use a commercial classifier; that is what they are calibrated for.
- The draft is shorter than a hundred words. The signal-to-noise ratio is too low; iterate by hand.

---

## Roadmap

- Closing-session capture for editors beyond Claude Code (the Claude Code hook ships in `callus.hooks`)
- Semantic-embeddings similarity backend as an optional `callus[embeddings]` extra — the lexical similarity layer ships now (`callus similarity`); a semantic backend drops in via `CALLUS_SIMILARITY_BACKEND`
- Multilingual corpus mixing rules (current default is single-language per corpus)

---

## Related projects

Part of a small cluster for operating LLM coding agents in production:

- **[lucy-syndrome](https://github.com/VDP89/lucy-syndrome)** — research on cross-session correction persistence (five invariants).
- **[fscars](https://github.com/VDP89/fscars)** — deterministic correction hooks (functional scars) for AI coding agents.

---

## Contributing

Issues and pull requests welcome. The interesting work right now is on calibrating the F-filters for other languages and on writing more eval sets so the rewriter's convergence behavior can be measured across more domains.

```bash
git clone https://github.com/VDP89/callus
cd callus
pip install -e ".[dev]"
pytest -q
```

---

## License

Apache 2.0 — see [LICENSE](LICENSE).
