semanticembed 0.7.2__tar.gz → 0.7.3__tar.gz

semanticembed-0.7.3/.github/ISSUE_TEMPLATE/bug_report.yml

@@ -0,0 +1,67 @@
+name: Bug report
+description: Report a bug in the SemanticEmbed SDK (parsers, client, live connectors, agent CLI).
+labels: ["bug"]
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Thanks for taking the time to report a bug. Please fill in the fields below.
+        For algorithm/encoding/scoring questions, email jeffmurr@seas.upenn.edu — that
+        runs server-side and isn't part of this repository.
+
+  - type: input
+    id: version
+    attributes:
+      label: SDK version
+      description: Output of `python -c "import semanticembed; print(semanticembed.__version__)"`
+      placeholder: "0.7.2"
+    validations:
+      required: true
+
+  - type: input
+    id: python
+    attributes:
+      label: Python version
+      placeholder: "3.12"
+    validations:
+      required: true
+
+  - type: dropdown
+    id: surface
+    attributes:
+      label: Which part of the SDK?
+      options:
+        - encode / aencode (HTTP client)
+        - extract.from_* (deterministic parsers)
+        - live.from_* (Dynatrace / Honeycomb / Datadog)
+        - find_edges (LLM fallback)
+        - dedupe_edges
+        - semanticembed-agent / semanticembed-gemini-agent CLI
+        - explain / ask
+        - other
+    validations:
+      required: true
+
+  - type: textarea
+    id: reproduce
+    attributes:
+      label: Minimal reproduction
+      description: Smallest snippet that triggers the bug. Please include the input edges or the path to the file you're parsing.
+      render: python
+    validations:
+      required: true
+
+  - type: textarea
+    id: expected
+    attributes:
+      label: Expected behavior
+    validations:
+      required: true
+
+  - type: textarea
+    id: actual
+    attributes:
+      label: Actual behavior (with traceback)
+      render: shell
+    validations:
+      required: true

semanticembed-0.7.3/.github/ISSUE_TEMPLATE/config.yml

@@ -0,0 +1,10 @@
+blank_issues_enabled: false
+contact_links:
+  - name: Algorithm / encoding / scoring questions
+    url: mailto:jeffmurr@seas.upenn.edu
+    about: |
+      The 6D encoding algorithm runs server-side and isn't part of this repository.
+      Email directly for questions about encoding math, risk thresholds, or scoring rules.
+  - name: License / commercial inquiries
+    url: mailto:jeffmurr@seas.upenn.edu
+    about: Pricing, paid-tier license keys, and commercial use questions.

semanticembed-0.7.3/.github/ISSUE_TEMPLATE/feature_request.yml

@@ -0,0 +1,48 @@
+name: Feature request
+description: Request a new parser, live connector, or SDK improvement.
+labels: ["enhancement"]
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Best fits for new features:
+        - **New `extract.from_*` parsers** — additional infrastructure formats, AI agent frameworks
+        - **New `live.from_*` connectors** — observability platforms (Lightstep, AppDynamics, New Relic, etc.)
+        - **Client improvements** — async, batching, caching, error handling
+
+        For changes to the encoding algorithm, risk thresholds, or scoring rules, please email
+        jeffmurr@seas.upenn.edu instead — those run server-side.
+
+  - type: dropdown
+    id: kind
+    attributes:
+      label: What kind of feature?
+      options:
+        - New extract.from_* parser
+        - New live.from_* connector
+        - Async / batching / caching improvement
+        - Error handling / retry behavior
+        - Documentation / example
+        - Other
+    validations:
+      required: true
+
+  - type: textarea
+    id: use_case
+    attributes:
+      label: Use case
+      description: What problem are you trying to solve? What's blocking you today?
+    validations:
+      required: true
+
+  - type: textarea
+    id: api_sketch
+    attributes:
+      label: Proposed API (optional)
+      description: A rough sketch of how you'd want to call it.
+      render: python
+
+  - type: textarea
+    id: alternatives
+    attributes:
+      label: Alternatives you've considered

semanticembed-0.7.3/.github/workflows/ci.yml

@@ -0,0 +1,59 @@
+name: CI
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+
+jobs:
+  test:
+    name: Test (${{ matrix.python-version }} on ${{ matrix.os }})
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        os: [ubuntu-latest]
+        python-version: ["3.10", "3.11", "3.12", "3.13"]
+        include:
+          # Sanity check on Windows + macOS at one Python version
+          - os: windows-latest
+            python-version: "3.12"
+          - os: macos-latest
+            python-version: "3.12"
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v5
+        with:
+          python-version: ${{ matrix.python-version }}
+          cache: pip
+
+      - name: Install
+        run: |
+          python -m pip install --upgrade pip
+          pip install -e '.[dev,extract]'
+
+      - name: Run pytest
+        run: pytest -q
+
+  build:
+    name: Build wheel + sdist
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: actions/setup-python@v5
+        with:
+          python-version: "3.12"
+
+      - name: Install build tooling
+        run: python -m pip install --upgrade build twine
+
+      - name: Build
+        run: python -m build
+
+      - name: Twine check (verifies README renders on PyPI)
+        run: python -m twine check dist/*

semanticembed-0.7.3/.gitignore

@@ -0,0 +1,22 @@
+__pycache__/
+*.pyc
+*.pyo
+dist/
+build/
+*.egg-info/
+.eggs/
+.env
+.venv/
+*.egg
+.ipynb_checkpoints/
+
+# Local-only artifacts at repo root (canonical files are in docs/ or examples/)
+/dimensions.md
+/encoding_output.json
+/*.html
+/*.pdf
+
+# Editor / OS noise
+.idea/
+.vscode/
+.DS_Store

{semanticembed-0.7.2 → semanticembed-0.7.3}/CHANGELOG.md

@@ -2,6 +2,25 @@
 
 All notable changes to the SemanticEmbed SDK.
 
+## 0.7.3 — Modern AutoGen + lenient edge parser
+
+- `from_autogen` now recognizes the autogen-agentchat 0.4+ patterns:
+  `RoundRobinGroupChat([...])` (chain), `SelectorGroupChat([...])` (fully
+  connected), `Swarm([...])` (fully connected), and
+  `MagenticOneGroupChat([...])`. The legacy `GroupChat` + optional
+  `GroupChatManager` paths still work.
+- `from_directory` auto-detects autogen-agentchat / autogen-core /
+  autogen-ext imports as well as the legacy `autogen` and `ag2` names.
+- HF Space (`huggingface/app.py`):
+  - Edge-list parser now tolerates BOM, smart quotes, tabs / pipes /
+    semicolons as delimiters, comment lines, and arrow-list syntax
+    (`a -> b`, `a => b`, `a → b`).
+  - "No edges extracted" error is now mode-specific: lists exactly which
+    patterns the LangGraph / CrewAI / AutoGen parser detects, with a
+    pointer to the Edge-list mode as the fallback.
+- README + Space landing copy: "Live dashboard" → "Demo dashboard"
+  (more accurate framing — it's a demo, not a production tool).
+
 ## 0.7.2 — Live-connector retry
 
 - All three live connectors (`from_dynatrace`, `from_honeycomb`, `from_datadog`)

semanticembed-0.7.3/CODE_OF_CONDUCT.md

@@ -0,0 +1,43 @@
+# Code of Conduct
+
+This project follows the [Contributor Covenant v2.1](https://www.contributor-covenant.org/version/2/1/code_of_conduct/).
+
+## Our Pledge
+
+We pledge to make participation in this project a harassment-free
+experience for everyone, regardless of age, body size, visible or
+invisible disability, ethnicity, sex characteristics, gender identity
+and expression, level of experience, education, socio-economic status,
+nationality, personal appearance, race, religion, or sexual identity
+and orientation.
+
+## Standards
+
+Examples of behavior that contributes to a positive environment:
+
+- Demonstrating empathy and kindness toward others
+- Being respectful of differing opinions and viewpoints
+- Giving and gracefully accepting constructive feedback
+- Accepting responsibility for mistakes and learning from them
+
+Examples of unacceptable behavior:
+
+- Trolling, insulting, derogatory comments, or personal attacks
+- Public or private harassment
+- Publishing others' private information without permission
+- Other conduct that could reasonably be considered inappropriate
+
+## Enforcement
+
+Instances of unacceptable behavior may be reported to the maintainer at
+**jeffmurr@seas.upenn.edu**. All complaints will be reviewed and
+investigated promptly and fairly.
+
+The maintainer is obligated to respect the privacy and security of the
+reporter of any incident.
+
+## Scope
+
+This Code of Conduct applies within all project spaces, including the
+GitHub repository, issue tracker, pull requests, discussions, and any
+public communication where an individual is representing the project.

semanticembed-0.7.3/CONTRIBUTING.md

@@ -0,0 +1,70 @@
+# Contributing
+
+Thanks for considering a contribution. The SDK is the public face of a
+proprietary product, so contribution scope is intentionally narrow.
+
+## What we welcome
+
+- **New edge extractors** in `extract.py` — additional infrastructure formats,
+  AI agent frameworks, CI/CD systems. Pure Python, ideally AST or text
+  parsing without runtime dependencies on the framework being parsed.
+- **Live observability connectors** in `live.py` — third-party APIs that
+  expose service-to-service call relationships (Lightstep, AppDynamics,
+  New Relic, Grafana Tempo, Sentry, etc.).
+- **Bug fixes** in `client.py`, `extract.py`, `live.py`, `dedupe.py`,
+  `find_edges.py`.
+- **Documentation** — README clarifications, notebook improvements, new
+  example fixtures in `examples/`.
+- **Test coverage** — additional respx-mocked test cases for edge cases.
+
+## What's out of scope
+
+- The 6D encoding algorithm itself runs server-side and is not part of
+  this repository. Any PR proposing changes to encoding math, risk
+  thresholds, or scoring rules will be closed.
+- Source-code modifications to the bundled console scripts that change
+  their published behavior require a coordinated release.
+- License changes. The SDK is proprietary; the public source code is for
+  adoption velocity, not distribution rights.
+
+## Development setup
+
+```bash
+git clone https://github.com/jmurray10/semanticembed-sdk
+cd semanticembed-sdk
+pip install -e '.[dev,extract]'
+pytest
+```
+
+The dev extras pull in `pytest`, `respx`, and `pyyaml`. All HTTP calls in
+the test suite are respx-mocked — no live network or credentials needed.
+
+## Pull-request checklist
+
+1. **Tests** — every new public function ships with at least one test in
+   `tests/`. Use `respx` for any HTTP, never call live APIs.
+2. **Type hints** — public APIs annotate arguments and return types.
+3. **Docstrings** — public functions have a docstring with a usage
+   `Example::` block.
+4. **CHANGELOG.md** — add an entry under an "Unreleased" header (we'll
+   roll it into the next version on merge).
+5. **No new runtime deps** without prior discussion. The SDK keeps a
+   minimal footprint (currently just `httpx`).
+
+## Releasing (maintainers)
+
+Tag pushes to `v*` trigger the `publish.yml` workflow, which uses PyPI
+Trusted Publishing (no token to manage). Steps:
+
+```bash
+# bump version in pyproject.toml AND src/semanticembed/__init__.py
+git commit -am "vX.Y.Z: description"
+git push
+git tag -a vX.Y.Z -m "vX.Y.Z — short summary"
+git push origin vX.Y.Z
+# workflow builds + publishes within ~2 minutes
+```
+
+## Questions
+
+Open a GitHub Discussion or email jeffmurr@seas.upenn.edu.

semanticembed-0.7.3/LICENSE-FAQ.md

@@ -0,0 +1,106 @@
+# License FAQ
+
+Common questions about the SemanticEmbed SDK license. The license itself
+is in [LICENSE](LICENSE).
+
+---
+
+## Is this open source?
+
+**No.** The SDK is **proprietary software with public source code**. The
+source is on GitHub for adoption velocity (so you can read it, fork it
+locally, and integrate it without friction), but the license does not
+grant the rights an OSI-approved open-source license would.
+
+This is the same pattern Stripe, Snowflake (`snowflake-connector-python`),
+and Anthropic (TypeScript SDK) use for their official client libraries.
+It's "source-available", not "open source".
+
+---
+
+## Can I use it commercially?
+
+**Yes — within the free tier.** Graphs up to 50 nodes per request are
+free for any use including commercial. No signup, no API key required.
+
+For graphs larger than 50 nodes, you need a paid license. Email
+**jeffmurr@seas.upenn.edu** for pricing.
+
+---
+
+## Can I fork the repo? Modify the code?
+
+You can fork the repo and modify your local copy for your own internal
+use, debugging, integration, or to send a PR back. You can't redistribute
+modified versions, package a derivative SDK, or relicense the source.
+The upshot: hack on it freely; don't ship a fork as your own product.
+
+---
+
+## What's actually proprietary?
+
+Two things:
+
+1. **The 6D encoding algorithm.** It runs server-side at the SemanticEmbed
+   cloud API; it's not in this repo. Patent application
+   [#63/994,075](#patent) covers the encoding method.
+2. **The SDK code in this repo.** All rights reserved per LICENSE.txt;
+   "All rights reserved" means the standard copyright defaults, modified
+   only by the explicit grants in the license.
+
+The SDK is a thin HTTP client + edge parsers. Reading the SDK source
+gives you no information about the encoding algorithm — that's the whole
+point of the architecture.
+
+---
+
+## What if my company has a strict OSS-only policy?
+
+Talk to us. The SDK is proprietary but not adversarial — we've been
+through procurement reviews where the legal team's primary concern was
+"is this going to disappear?" (answer: paid customers get an escrow
+clause; free-tier doesn't but everything is also pinned on PyPI). Email
+**jeffmurr@seas.upenn.edu**.
+
+---
+
+## <a name="patent"></a>Patent
+
+US Provisional Patent Application **#63/994,075**, filed
+**2 March 2026**, covers the 6D structural encoding method, the risk
+classification rules, and the system architecture (thin client + cloud
+encoding service).
+
+The patent application is in prosecution — it has not yet issued. The
+SDK is shipped and supported regardless of patent timing.
+
+---
+
+## Free tier specifics
+
+- **Hard limit:** 50 unique nodes per `encode()` call. Fewer than 2
+  edges raises `ValueError`.
+- **No rate limit on the wire** for the free path, but the SDK's
+  retry-once policy and the server's tier policy together cap practical
+  burst rate. If you're hitting rate limits, you're probably ready for
+  a paid license anyway.
+- **No signup, no email collection.** The free path doesn't even know
+  who you are.
+
+---
+
+## Paid tier
+
+- Larger graphs (configurable per license).
+- Continuous monitoring (drift detection, scheduled re-encodes).
+- CI/CD gating (block PRs that worsen structural risk past a threshold).
+- Priority support, SLA, on-prem deployment options.
+
+Pricing is bespoke depending on node count + monitoring frequency.
+Email **jeffmurr@seas.upenn.edu** with your use case.
+
+---
+
+## More questions
+
+Anything not covered here: **jeffmurr@seas.upenn.edu**.