Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

OpenSourceScout

A decision framework for scouting GitHub repos (and ideas) as a solo builder working with agentic AI. Point it at a repo; it tells you whether to build on it independently, fold it into an existing project, harvest parts of it, save it as inventory for later, or pass — and why.

It is not a scorer. It routes.

The model — Assess → Route → Capital

ASSESS   facts + worthiness, in one pass
           facts:      category · payoff · features · ai_at_build/run · license · liability
           worthiness: ≥1 payoff  +  passes your standards  +  warrants your time  +  liability LOW
                       │
ROUTE      independent → too_big → merge → combined → rejected
                       │
CAPITAL    launch must be free (else parked_capital); scale from revenue
                       → status: active | parked_capital
  • independent — worth doing AND within one-person-plus-AI scope → build it.
  • too_big — worth doing but needs a team → defer / partner.
  • merge — folds into an existing project: as_is · modify_fully · harvest_parts.
  • combined — its features help cover a target when pooled with other repos.
  • rejected — retained as feature-tagged inventory (supply and inspiration for future targets).

Guiding rules: never reject on spoken/programming language or on capital; AI-ness is a category tag, never a merit; money is the last gate, never the first. Full rules in criteria.yaml; design in SPEC.md.

Repo track (start here)

You feed it a repo; it gathers GitHub metadata, runs the funnel, and writes a verdict note. The instructions the judge follows are in prompts/repo-judge.md.

Drop repos to evaluate (one owner/name or URL per line) into inbox/, or pass them directly to the judge.

Layout

criteria.yaml            the shareable rules (public framework)
SPEC.md                  design + rationale
prompts/repo-judge.md    how the repo track evaluates a candidate
knowledge/repos/         verdict notes  (PRIVATE — git-ignored; _TEMPLATE.md is tracked)
knowledge/targets/       wish-list products to assemble toward (PRIVATE)
knowledge/{topics,daily} rollups + run logs (PRIVATE)
config.local.yaml        your portfolio, standards, categories (PRIVATE — git-ignored)

Public framework vs. private data. The rules, spec, prompts, and templates are shareable. Your portfolio, the repos you're weighing, and your verdicts are not — they live in git-ignored paths (see .gitignore). Copy config.local.yaml and fill it in before first use.

Status: framework complete; repo track ready to run. Idea track is planned.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors