Open Source CLI

Deterministic.
Auditable.
Cost-efficient.

Tangent sits between your spec and your coding agent — constraining AI generation through opinionated templates while staying intelligent exactly where it needs to be.

View on GitHub
pip install tangent-cli
Templates alone

Predictable, reviewable — but too rigid to express real logic.

Too rigid
tangentThe synthesis
AI agents alone

Flexible, fast — but unauditable, non-reproducible, and expensive.

Too loose

How Tangent works

Four primitives that turn AI code generation into a repeatable engineering process.

  1. 01

    Jinja templates with surgical LLM calls

    Write templates that embed LLM calls only where static output can't express the logic. Everything else is deterministic by construction.

  2. 02

    DAG resolves dependencies

    Tangent builds a dependency graph across all LLM calls so outputs are generated in the correct order, every time.

  3. 03

    Content-addressed lockfile

    Every result is hashed and stored. Only inputs that actually changed trigger a new LLM call — the rest is served from the cache.

  4. 04

    Tamper-evident audit trail

    The lockfile records exactly what was generated, from what inputs, with what model. Commit it. Review it. Diff it like any other source artifact.

95%

reduction in AI inference costs on repeat runs — because unchanged inputs never re-invoke the model.

Built for every scale

The same tool that helps a solo hacker ship fast also satisfies an enterprise compliance team.

Solo developers & indie hackers

Ship fast with AI and actually be able to maintain what you shipped. Every generated file conforms to the architectural patterns you reviewed once — not whatever the agent felt like that day.

Engineering leads

Replace unauditable AI output with traceable, reproducible generation. Review architectural decisions once at the template level. Every subsequent generation is consistent by construction.

Enterprise teams

The lockfile is a tamper-evident audit trail. Planned policy enforcement constrains filesystem, network, and dependency usage — containing the compliance risk that unconstrained generative code creates.

Ready to make your AI generation trustworthy?

Open source. Designed to fit into your existing workflow.

Get started on GitHub