Skip to content

Repository files navigation

RevEng

Inspect an unfamiliar Python repository and get source-linked structural maps, reports and explicit unknowns.

Try it

Python 3.11+. Run from this checkout:

python -m pip install -e .
python -m examples.offline_demo --out output/first-analysis

Input: Three small Python files with explicit imports and function calls.

Result: The static workflow writes a file inventory, relationship map, per-file breakdowns, dossier, validation and unknowns. It does not execute the input repository or call a model.

See the captured example for the observed output and reproduction command. Then follow the source-to-result walkthrough: main.py:5 becomes a call edge, six consistency checks pass, and four static-analysis caveats remain visible.

How it works

An inspectable architecture account starts from source symbols and relationships, then carries evidence into derived reports. Explicit unknowns show where static analysis cannot resolve behavior.

Source: reveng/cli.py, reveng/analysis_engine/workflows/repo_analysis.py, tests/test_host_workflow.py.

Use it for your work

Run reveng for the local browser interface, then choose a Python repository you want to understand. Start with the included examples/tiny_repository; output folders must be new. Prefer a short runtime path on Windows. See ASSETS.md for interface-asset provenance.

Beyond the first analysis

The local control center also manages reusable capabilities, composite workflows, agent permissions, run evidence and publication history. These are implemented surfaces with their own contracts; the first static analysis does not require configuring them. ARCHITECTURE.md maps the components and SECURITY.md explains the local HTTP boundary and trusted authored-code setting.

Keep the default loopback binding. Optional user-authored Python runs with the host process’s access after explicit enablement; capability keycards are not an operating-system sandbox.

Scope

Static relationships and inferred behavior are provisional. Dynamic imports/calls and framework behavior may remain unresolved. Optional model/desktop surfaces need their own configuration; the first example uses neither.

Owned code is available under the MIT license.

About

Inspect an unfamiliar Python repository and get source-linked structural maps, reports and explicit unknowns.

Topics

Resources

Security policy

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages