How I Built RepoSéance: A Local-First AI Ghost Hunter That Resurrects Abandoned Codebases

Description and use

RepoSéance is a local-first tool that scans abandoned repositories, builds an offline vector index, and uses open-weight LLMs to reconstruct architecture, identify entry points, and generate a revival roadmap, all while keeping data private.

Problem

Developers often encounter abandoned GitHub repositories with unclear entry points, deprecated dependencies, and undocumented assumptions, making it hard to assess their viability or revive them while preserving privacy.

Outcome

RepoSéance transforms a neglected codebase into a documented, analyzable artifact with a clear entry point and actionable tasks, all while keeping data local and private.

Workflow

• Select local folder or ZIP
• Scanner validates path, respects .gitignore, redacts credentials, extracts Git history
• Local embeddings built with nomic-embed-text or bge-small-en
• Vector index created
• Open‑weight LLM (Qwen2.5-Coder, Qwen2.5-3B, Gemma 2-2B via Ollama) performs structured autopsy
• Ghost Chat answers queries with file‑path citations
• Revival roadmap generated and exported as Markdown

Evidence

RepoSéance is a local-first, privacy-respecting software archaeology companion. You provide a local directory path or upload a ZIP archive of any neglected project, and RepoSéance conducts a digital séance:

Source

Deterministic Repository Archaeology: Safely maps directory structures, scans dependencies, inspects local Git commit histories, and catalogs buried TODO/FIXME markers without executing a single line of untrusted code.

Source

Local Code Indexing & Bounded Evidence Retrieval: Chunks source code, extracts semantic tokens, and builds an offline vector index using local embedding models (nomic-embed-text / bge-small-en).

Source

Structured Project Autopsy: Uses a locally running open-weight model (Qwen2.5-Coder / Qwen2.5-3B / Gemma 2-2B via Ollama) to reconstruct the project's true purpose, identify architecture flows, highlight entry points, and distinguish verified facts from AI hypotheses.

Source

RepoSéance requires zero external API keys and runs completely on standard laptop hardware.

Source