dont-break 0.3.1__tar.gz

dont_break-0.3.1/.env.example

@@ -0,0 +1,18 @@
+# Optional. End users need none of this — `dont-break --wake` talks to production.
+#
+# To override (local gateway / staging), put the same keys in:
+#   ~/.config/dont-break/.env
+# or export them in your shell. A cwd `.env.local` is intentionally ignored.
+
+# POLYMERIX_API_BASE_URL=http://localhost:8080
+# POLYMERIX_APP_URL=http://localhost:3000
+
+# Optional workspace override (defaults to org_slug from sign-in)
+# POLYMERIX_ORG_SLUG=polymerix-labs
+
+# Local UI bind (defaults: 127.0.0.1:4040)
+# DONT_BREAK_HOST=127.0.0.1
+# DONT_BREAK_PORT=4040
+
+# Front: @polymerix-labs/facts-extract (npm). Override if not using npx / node_modules.
+# POLYMERIX_FACTS_EXTRACT_EXECUTABLE=/path/to/facts-extract

dont_break-0.3.1/.gitignore

@@ -0,0 +1,19 @@
+__pycache__/
+*.py[cod]
+.venv*/
+venv/
+.env
+.env.local
+dist/
+/build/
+*.egg-info/
+.dont-break/
+
+frontend/node_modules/
+frontend/dist/
+frontend/tsconfig.tsbuildinfo
+dont_break/static/
+
+.DS_Store
+node_modules/
+frontend/dont_break/static/

dont_break-0.3.1/LICENSE

@@ -0,0 +1,177 @@
+
+                                 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

dont_break-0.3.1/NOTICE

@@ -0,0 +1,4 @@
+dont-break
+Copyright 2026 Polymerix
+
+This product includes software developed by Polymerix (https://dont-break.com).

dont_break-0.3.1/PKG-INFO

@@ -0,0 +1,141 @@
+Metadata-Version: 2.5
+Name: dont-break
+Version: 0.3.1
+Summary: The trust layer for AI-written code: say what must never break, prove the protection holds, enforce it on every agent edit.
+Project-URL: Homepage, https://dont-break.com
+Project-URL: Documentation, https://dont-break.com/docs
+Project-URL: Source, https://github.com/polymerix-labs/dont-break
+Project-URL: Issues, https://github.com/polymerix-labs/dont-break/issues
+Author-email: Polymerix <hello@dont-break.com>
+License-Expression: Apache-2.0
+License-File: LICENSE
+License-File: NOTICE
+Keywords: ai-agents,claude,code-graph,code-review,cursor,dependency-graph,impact-analysis,mcp,refactoring,static-analysis
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Operating System :: OS Independent
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+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 :: Build Tools
+Classifier: Topic :: Software Development :: Quality Assurance
+Requires-Python: >=3.9
+Requires-Dist: fastapi>=0.115.0
+Requires-Dist: httpx>=0.27.0
+Requires-Dist: pydantic-settings>=2.6.0
+Requires-Dist: rich>=13.9.0
+Requires-Dist: uvicorn[standard]>=0.32.0
+Requires-Dist: watchfiles>=0.24
+Requires-Dist: websockets>=14.0
+Requires-Dist: zstandard>=0.23
+Provides-Extra: dev
+Requires-Dist: httpx>=0.27; extra == 'dev'
+Requires-Dist: pytest-asyncio>=0.24; extra == 'dev'
+Requires-Dist: pytest>=8.0; extra == 'dev'
+Requires-Dist: ruff>=0.8; extra == 'dev'
+Description-Content-Type: text/markdown
+
+# dont-break
+
+**The trust layer for AI-written code.**
+
+[![PyPI](https://img.shields.io/pypi/v/dont-break)](https://pypi.org/project/dont-break/)
+[![Python](https://img.shields.io/pypi/pyversions/dont-break)](https://pypi.org/project/dont-break/)
+[![Downloads](https://img.shields.io/pypi/dm/dont-break)](https://pypi.org/project/dont-break/)
+[![License](https://img.shields.io/badge/license-Apache--2.0-blue)](https://github.com/polymerix-labs/dont-break/blob/main/LICENSE)
+
+English · [Français](https://github.com/polymerix-labs/dont-break/blob/main/README.fr.md)
+
+![Rule Studio: describe what must never break, watch the graph find and test the protection](https://raw.githubusercontent.com/polymerix-labs/dont-break/main/docs/assets/screenshot.png)
+
+AI agents ship code fast. Nobody ships trust with it. Every team running Cursor, Claude, or CI bots has the same unspoken fear: the day a quick fix silently breaks the one thing that was never supposed to break.
+
+`dont-break` turns that fear into a contract:
+
+1. **Say it in plain words.** "Nobody should be able to break invoice calculation, even indirectly." No file paths, no code.
+2. **Watch it get found.** dont-break reads the live map of your codebase and lights up every place that carries that logic, including the paths you forgot existed.
+3. **Watch it get attacked.** It writes a protection rule, then replays code changes against it to prove the protection actually holds. A dry run: nothing in your code is touched.
+4. **Activate it.** From then on, every agent edit is checked against the rule before it lands. Your agent gets told "this is riskier than it looks" instead of you finding out in prod.
+
+```text
+You:    "Rename PokemonService.fetchAll"
+Agent:  → get_dependents(PokemonService.fetchAll)   "23 call sites across 4 modules"
+        → get_impact(files: [...])                  "radius 3, touches ui/, cache/, api/"
+        → get_do_not_touch()                        "PokemonService is a danger zone: fan-in 23, stability 31"
+Agent:  "This is riskier than it looks. Here are the 23 places that break,
+         and a safer 2-step plan."
+```
+
+That conversation happens automatically once connected. No prompt engineering: the agent skill teaches it.
+
+## Install
+
+Requires **Python 3.9+** and **Node.js** (npm). The graph extractor installs itself on first run.
+
+```bash
+pip install dont-break
+dont-break --wake
+```
+
+That opens a local UI on `http://127.0.0.1:4040`, in your language (30 available). Sign in, pick a project folder, and the map of your code builds itself: a live 3D graph of every module, call, and dependency, with your protected zones lit up on top of it.
+
+## Pick your fight
+
+**"My agent keeps breaking things it never opened"**
+Plug dont-break into Cursor or Claude Desktop. Your agent checks impact and danger zones before editing, not after.
+→ [Set up in Cursor / Claude (2 min)](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/cursor.md)
+
+**"I want CI to block disasters, not argue about style"**
+One job that fails the merge when a change hits a protected zone or a fragile hub, grounded in the real dependency graph, not vibes.
+→ [GitHub Actions](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/github-actions.md) · [GitLab CI](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/gitlab-ci.md) · [pre-commit hook](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/pre-commit.md)
+
+**"I just want to interrogate my codebase"**
+`dbq dependents <id> | jq`: who breaks if I change this? Your repo becomes a queryable database.
+→ [Shell + jq recipes](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/shell.md)
+
+**"I'm building my own agent"**
+The same 11 tools, exposed as typed TypeScript definitions or a generated OpenAPI 3.1 spec.
+→ [LangChain / OpenAPI / custom agents](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/langchain.md)
+
+## The 11 tools your agent gets
+
+| Tool | The question it kills |
+|------|----------------------|
+| `find_symbol` | "Which node is this name / file?" (entry point) |
+| `get_dependents` | "Who breaks if I change this?" |
+| `get_impact` | "What's the blast radius of these edits?" |
+| `get_do_not_touch` | "What should I refuse to touch without asking?" |
+| `get_dependencies` | "What does this code rely on?" |
+| `find_path` | "Why does changing A affect B?" |
+| `get_arch_status` | "How carefully should I work in this repo?" |
+| `check_change` | "Does this edit violate a team rule?" |
+| `propose_rule` | "Record a warn now, or a block for a human to approve" |
+| `pause_own_rule` | "Stop evaluating a rule this agent token authored" |
+| `append_rule_reason` | "Add one justification, never edit or delete" |
+
+Query tools are read-only, server-side analysis, capped responses: always safe to call. The three rule tools write team rules under tight limits: they cannot activate a block, pause someone else's rule, or rewrite reasons.
+
+## The control room
+
+- **Rule Studio**: describe what must never break, watch the graph find it, test the protection live before activating it
+- **Check**: pre-edit simulator: pick seeds, get an ok/warn/block verdict, animate the exact path a break would take
+- **Overview**: a verdict in one sentence, stability and AI-navigability readouts, the top actions that would harden your architecture
+- **Graph**: the Nebula 3D scene, protected zones and witness paths lit as overlays
+- **Agents**: connect Cursor, Claude, or CI in one click, with a live try-to-break demo
+
+Keyboard-first: `cmd+K` opens the command palette.
+
+## Connect your agent in 30 seconds
+
+1. Open the dont-break app → **Agents**.
+2. Sign in, link the folder to a project, click **Connect Cursor**: one click mints a project-scoped token and fills `mcp.json`.
+3. Paste into Cursor (or your MCP client).
+4. Click **Install agent skill**: it writes the safe-change protocol into your repo's `AGENTS.md` so agents use the tools without being told.
+
+## License
+
+Apache-2.0. See [LICENSE](https://github.com/polymerix-labs/dont-break/blob/main/LICENSE) and [NOTICE](https://github.com/polymerix-labs/dont-break/blob/main/NOTICE).

dont_break-0.3.1/README.fr.md

@@ -0,0 +1,100 @@
+# dont-break
+
+**La couche de confiance du code écrit par IA.**
+
+[![PyPI](https://img.shields.io/pypi/v/dont-break)](https://pypi.org/project/dont-break/)
+[![Python](https://img.shields.io/pypi/pyversions/dont-break)](https://pypi.org/project/dont-break/)
+[![Downloads](https://img.shields.io/pypi/dm/dont-break)](https://pypi.org/project/dont-break/)
+[![License](https://img.shields.io/badge/license-Apache--2.0-blue)](https://github.com/polymerix-labs/dont-break/blob/main/LICENSE)
+
+[English](https://github.com/polymerix-labs/dont-break/blob/main/README.md) · Français
+
+![Rule Studio : dites ce qui ne doit jamais casser, regardez le graphe trouver et tester la protection](https://raw.githubusercontent.com/polymerix-labs/dont-break/main/docs/assets/screenshot.png)
+
+Les agents IA livrent du code vite. Personne ne livre la confiance avec. Toutes les équipes qui utilisent Cursor, Claude ou des bots CI partagent la même crainte : le jour où un correctif rapide casse silencieusement la seule chose qui ne devait jamais casser.
+
+`dont-break` transforme cette crainte en contrat :
+
+1. **Dites-le avec des mots simples.** « Personne ne doit pouvoir casser le calcul des factures, même indirectement. » Pas de chemins de fichiers, pas de code.
+2. **Regardez-le être trouvé.** dont-break lit la carte vivante de votre code et illumine chaque endroit qui porte cette logique, y compris les chemins que vous aviez oubliés.
+3. **Regardez-le être attaqué.** Il écrit une règle de protection, puis rejoue des changements de code contre elle pour prouver qu'elle tient vraiment. Un test à blanc : rien dans votre code n'est modifié.
+4. **Activez-la.** Ensuite, chaque édition d'agent est vérifiée contre la règle avant d'atterrir. Votre agent s'entend dire « c'est plus risqué qu'il n'y paraît » au lieu que vous le découvriez en prod.
+
+```text
+Vous :   « Renomme PokemonService.fetchAll »
+Agent :  → get_dependents(PokemonService.fetchAll)   « 23 appels dans 4 modules »
+         → get_impact(files: [...])                  « rayon 3, touche ui/, cache/, api/ »
+         → get_do_not_touch()                        « PokemonService est une zone dangereuse : fan-in 23, stabilité 31 »
+Agent :  « C'est plus risqué qu'il n'y paraît. Voici les 23 endroits qui cassent,
+           et un plan plus sûr en 2 étapes. »
+```
+
+Cette conversation se produit automatiquement une fois connecté. Aucun prompt engineering : le skill d'agent l'enseigne.
+
+## Installation
+
+Nécessite **Python 3.9+** et **Node.js** (npm). L'extracteur de graphe s'installe tout seul au premier lancement.
+
+```bash
+pip install dont-break
+dont-break --wake
+```
+
+Une interface locale s'ouvre sur `http://127.0.0.1:4040`, dans votre langue (30 disponibles). Connectez-vous, choisissez un dossier projet, et la carte de votre code se construit toute seule : un graphe 3D vivant de chaque module, appel et dépendance, avec vos zones protégées illuminées par-dessus.
+
+## Choisissez votre combat
+
+**« Mon agent casse sans arrêt des choses qu'il n'a jamais ouvertes »**
+Branchez dont-break dans Cursor ou Claude Desktop. Votre agent vérifie l'impact et les zones dangereuses avant d'éditer, pas après.
+→ [Configuration Cursor / Claude (2 min)](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/cursor.md)
+
+**« Je veux une CI qui bloque les désastres, pas qui discute du style »**
+Un seul job qui fait échouer le merge quand un changement touche une zone protégée ou un module fragile, fondé sur le vrai graphe de dépendances.
+→ [GitHub Actions](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/github-actions.md) · [GitLab CI](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/gitlab-ci.md) · [hook pre-commit](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/pre-commit.md)
+
+**« Je veux juste interroger mon code »**
+`dbq dependents <id> | jq` : qui casse si je change ça ? Votre repo devient une base de données interrogeable.
+→ [Recettes shell + jq](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/shell.md)
+
+**« Je construis mon propre agent »**
+Les mêmes 11 outils, exposés en définitions TypeScript typées ou en spec OpenAPI 3.1 générée.
+→ [LangChain / OpenAPI / agents maison](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/langchain.md)
+
+## Les 11 outils de votre agent
+
+| Outil | La question qu'il élimine |
+|-------|---------------------------|
+| `find_symbol` | « Quel nœud correspond à ce nom / fichier ? » (point d'entrée) |
+| `get_dependents` | « Qui casse si je change ça ? » |
+| `get_impact` | « Quel est le rayon d'impact de ces modifications ? » |
+| `get_do_not_touch` | « Que dois-je refuser de toucher sans demander ? » |
+| `get_dependencies` | « De quoi ce code dépend-il ? » |
+| `find_path` | « Pourquoi changer A affecte B ? » |
+| `get_arch_status` | « Avec quelle prudence travailler dans ce repo ? » |
+| `check_change` | « Cette modification viole-t-elle une règle d'équipe ? » |
+| `propose_rule` | « Enregistrer un warn maintenant, ou un block à faire approuver » |
+| `pause_own_rule` | « Suspendre une règle créée par ce token d'agent » |
+| `append_rule_reason` | « Ajouter une justification, jamais modifier ni supprimer » |
+
+Les outils de requête sont en lecture seule, analyse côté serveur, réponses plafonnées : toujours sûrs à appeler. Les trois outils de règles écrivent des règles d'équipe sous limites strictes : ils ne peuvent pas activer un block, suspendre la règle de quelqu'un d'autre, ni réécrire des justifications.
+
+## La salle de contrôle
+
+- **Rule Studio** : décrivez ce qui ne doit jamais casser, regardez le graphe le trouver, testez la protection en direct avant de l'activer
+- **Check** : simulateur pré-édition : choisissez des points de départ, obtenez un verdict ok/warn/block, animez le chemin exact qu'une casse emprunterait
+- **Overview** : un verdict en une phrase, stabilité et navigabilité IA, les actions prioritaires pour durcir votre architecture
+- **Graph** : la scène 3D Nebula, zones protégées et chemins témoins illuminés en surcouche
+- **Agents** : connectez Cursor, Claude ou la CI en un clic, avec une démo « essayez de casser » en direct
+
+Piloté au clavier : `cmd+K` ouvre la palette de commandes.
+
+## Connectez votre agent en 30 secondes
+
+1. Ouvrez l'app dont-break → **Agents**.
+2. Connectez-vous, liez le dossier à un projet, cliquez **Connect Cursor** : un clic génère un token limité au projet et remplit `mcp.json`.
+3. Collez dans Cursor (ou votre client MCP).
+4. Cliquez **Install agent skill** : le protocole de modification sûre s'écrit dans l'`AGENTS.md` de votre repo, et les agents utilisent les outils sans qu'on le leur dise.
+
+## Licence
+
+Apache-2.0. Voir [LICENSE](https://github.com/polymerix-labs/dont-break/blob/main/LICENSE) et [NOTICE](https://github.com/polymerix-labs/dont-break/blob/main/NOTICE).

dont_break-0.3.1/README.md

@@ -0,0 +1,100 @@
+# dont-break
+
+**The trust layer for AI-written code.**
+
+[![PyPI](https://img.shields.io/pypi/v/dont-break)](https://pypi.org/project/dont-break/)
+[![Python](https://img.shields.io/pypi/pyversions/dont-break)](https://pypi.org/project/dont-break/)
+[![Downloads](https://img.shields.io/pypi/dm/dont-break)](https://pypi.org/project/dont-break/)
+[![License](https://img.shields.io/badge/license-Apache--2.0-blue)](https://github.com/polymerix-labs/dont-break/blob/main/LICENSE)
+
+English · [Français](https://github.com/polymerix-labs/dont-break/blob/main/README.fr.md)
+
+![Rule Studio: describe what must never break, watch the graph find and test the protection](https://raw.githubusercontent.com/polymerix-labs/dont-break/main/docs/assets/screenshot.png)
+
+AI agents ship code fast. Nobody ships trust with it. Every team running Cursor, Claude, or CI bots has the same unspoken fear: the day a quick fix silently breaks the one thing that was never supposed to break.
+
+`dont-break` turns that fear into a contract:
+
+1. **Say it in plain words.** "Nobody should be able to break invoice calculation, even indirectly." No file paths, no code.
+2. **Watch it get found.** dont-break reads the live map of your codebase and lights up every place that carries that logic, including the paths you forgot existed.
+3. **Watch it get attacked.** It writes a protection rule, then replays code changes against it to prove the protection actually holds. A dry run: nothing in your code is touched.
+4. **Activate it.** From then on, every agent edit is checked against the rule before it lands. Your agent gets told "this is riskier than it looks" instead of you finding out in prod.
+
+```text
+You:    "Rename PokemonService.fetchAll"
+Agent:  → get_dependents(PokemonService.fetchAll)   "23 call sites across 4 modules"
+        → get_impact(files: [...])                  "radius 3, touches ui/, cache/, api/"
+        → get_do_not_touch()                        "PokemonService is a danger zone: fan-in 23, stability 31"
+Agent:  "This is riskier than it looks. Here are the 23 places that break,
+         and a safer 2-step plan."
+```
+
+That conversation happens automatically once connected. No prompt engineering: the agent skill teaches it.
+
+## Install
+
+Requires **Python 3.9+** and **Node.js** (npm). The graph extractor installs itself on first run.
+
+```bash
+pip install dont-break
+dont-break --wake
+```
+
+That opens a local UI on `http://127.0.0.1:4040`, in your language (30 available). Sign in, pick a project folder, and the map of your code builds itself: a live 3D graph of every module, call, and dependency, with your protected zones lit up on top of it.
+
+## Pick your fight
+
+**"My agent keeps breaking things it never opened"**
+Plug dont-break into Cursor or Claude Desktop. Your agent checks impact and danger zones before editing, not after.
+→ [Set up in Cursor / Claude (2 min)](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/cursor.md)
+
+**"I want CI to block disasters, not argue about style"**
+One job that fails the merge when a change hits a protected zone or a fragile hub, grounded in the real dependency graph, not vibes.
+→ [GitHub Actions](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/github-actions.md) · [GitLab CI](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/gitlab-ci.md) · [pre-commit hook](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/pre-commit.md)
+
+**"I just want to interrogate my codebase"**
+`dbq dependents <id> | jq`: who breaks if I change this? Your repo becomes a queryable database.
+→ [Shell + jq recipes](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/shell.md)
+
+**"I'm building my own agent"**
+The same 11 tools, exposed as typed TypeScript definitions or a generated OpenAPI 3.1 spec.
+→ [LangChain / OpenAPI / custom agents](https://github.com/polymerix-labs/dont-break/blob/main/docs/agents/langchain.md)
+
+## The 11 tools your agent gets
+
+| Tool | The question it kills |
+|------|----------------------|
+| `find_symbol` | "Which node is this name / file?" (entry point) |
+| `get_dependents` | "Who breaks if I change this?" |
+| `get_impact` | "What's the blast radius of these edits?" |
+| `get_do_not_touch` | "What should I refuse to touch without asking?" |
+| `get_dependencies` | "What does this code rely on?" |
+| `find_path` | "Why does changing A affect B?" |
+| `get_arch_status` | "How carefully should I work in this repo?" |
+| `check_change` | "Does this edit violate a team rule?" |
+| `propose_rule` | "Record a warn now, or a block for a human to approve" |
+| `pause_own_rule` | "Stop evaluating a rule this agent token authored" |
+| `append_rule_reason` | "Add one justification, never edit or delete" |
+
+Query tools are read-only, server-side analysis, capped responses: always safe to call. The three rule tools write team rules under tight limits: they cannot activate a block, pause someone else's rule, or rewrite reasons.
+
+## The control room
+
+- **Rule Studio**: describe what must never break, watch the graph find it, test the protection live before activating it
+- **Check**: pre-edit simulator: pick seeds, get an ok/warn/block verdict, animate the exact path a break would take
+- **Overview**: a verdict in one sentence, stability and AI-navigability readouts, the top actions that would harden your architecture
+- **Graph**: the Nebula 3D scene, protected zones and witness paths lit as overlays
+- **Agents**: connect Cursor, Claude, or CI in one click, with a live try-to-break demo
+
+Keyboard-first: `cmd+K` opens the command palette.
+
+## Connect your agent in 30 seconds
+
+1. Open the dont-break app → **Agents**.
+2. Sign in, link the folder to a project, click **Connect Cursor**: one click mints a project-scoped token and fills `mcp.json`.
+3. Paste into Cursor (or your MCP client).
+4. Click **Install agent skill**: it writes the safe-change protocol into your repo's `AGENTS.md` so agents use the tools without being told.
+
+## License
+
+Apache-2.0. See [LICENSE](https://github.com/polymerix-labs/dont-break/blob/main/LICENSE) and [NOTICE](https://github.com/polymerix-labs/dont-break/blob/main/NOTICE).

dont_break-0.3.1/docs/agents/README.md

@@ -0,0 +1,69 @@
+# Your AI agent can't break what it can see
+
+AI agents edit code blind. They read the file in front of them, change a function, and have no idea that 47 other places depend on it. You find out in review — or in prod.
+
+**dont-break** gives any agent (Cursor, Claude, your CI, your own bot) a live structural map of your codebase: who depends on what, the blast radius of every change, and which parts are too fragile to touch. Before the edit, not after.
+
+```text
+You:    "Rename PokemonService.fetchAll"
+Agent:  → get_dependents(PokemonService.fetchAll)   "23 call sites across 4 modules"
+        → get_impact(files: [...])                  "radius 3, touches ui/, cache/, api/"
+        → get_do_not_touch()                        "PokemonService is a danger zone: fan-in 23, stability 31"
+Agent:  "This is riskier than it looks. Here are the 23 places that break,
+         and a safer 2-step plan."
+```
+
+That conversation happens automatically once connected. No prompt engineering required — the agent skill teaches it.
+
+## Pick your fight
+
+### "My agent keeps breaking things it never opened"
+
+Vibe coding is great until a quick fix ripples somewhere invisible. Plug dont-break into **Cursor or Claude Desktop** and your agent checks impact and danger zones *before* editing — then warns you instead of apologizing after.
+
+→ [Set up in Cursor / Claude (2 min)](cursor.md)
+
+### "I want CI to block disasters, not argue about style"
+
+Your linter can't see architecture. Add one job that fails the merge when the change hits a fragile hub, or just posts the real blast radius on every PR — grounded in the dependency graph, not vibes.
+
+→ [GitHub Actions](github-actions.md) · [GitLab CI](gitlab-ci.md) · [pre-commit hook](pre-commit.md)
+
+### "I just want to interrogate my codebase"
+
+`dbq dependents <id> | jq` — who breaks if I change this? What's the safest path between these two modules? Which files should nobody touch this sprint? Your repo becomes a queryable database.
+
+→ [Shell + jq recipes](shell.md)
+
+### "I'm building my own agent"
+
+Same 11 tools, exposed as typed TypeScript definitions or a generated OpenAPI 3.1 spec. Bind them to LangChain, GPT Actions, or anything that speaks function calling — identical behavior everywhere, because every surface derives from one core.
+
+→ [LangChain / OpenAPI / custom agents](langchain.md)
+
+## The 11 tools your agent gets
+
+| Tool | The question it kills |
+|------|----------------------|
+| `find_symbol` | "Which node is this name / file?" (entry point) |
+| `get_dependents` | "Who breaks if I change this?" |
+| `get_impact` | "What's the blast radius of these edits?" |
+| `get_do_not_touch` | "What should I refuse to touch without asking?" |
+| `get_dependencies` | "What does this code rely on?" |
+| `find_path` | "Why does changing A affect B?" |
+| `get_arch_status` | "How carefully should I work in this repo?" |
+| `check_change` | "Does this edit violate a team rule?" |
+| `propose_rule` | "Record a warn now, or a block for a human to approve" |
+| `pause_own_rule` | "Stop evaluating a rule this agent token authored" |
+| `append_rule_reason` | "Add one justification — never edit or delete" |
+
+Query tools are read-only, server-side analysis, capped responses. Calling them is always safe. The three rule tools write team rules under tight limits: they cannot activate a block, pause someone else's rule, or rewrite reasons.
+
+## Connect in 30 seconds
+
+1. Open the dont-break app → **Agents**.
+2. Sign in, link the folder to a project, click **Connect Cursor** — one click mints a project-scoped `dbt_` token and fills `mcp.json`.
+3. Paste into Cursor (or your MCP client). Copy it now: the secret is shown once.
+4. Click **Install agent skill** — it writes the safe-change protocol into your repo's `AGENTS.md` so agents use the tools without being told.
+
+Everything is configured through two environment variables (`DONT_BREAK_API_URL`, `DONT_BREAK_TOKEN`). The `dbt_` token binds the workspace and project server-side; tokens travel via environment only — never in argv, never in shell history. Manage / revoke / Sync & CI presets on the website **Account → API tokens** page.