tesserakit-repo

Map a repository into a validated structural artifact: a file inventory, a language and layout map, the declared dependency surface, and repo-hygiene signals.

tessera-repo walks a repository (ignoring build and vendor directories), classifies every file by language and kind, detects and parses dependency manifests, and emits a structural map plus reports. It reads code; it does not run it, and it makes no network calls.

Scope (v0.1)

Pure structural analysis. No code execution, no imports of the target repo, no network. The output is an offline, reviewable map of what a repository contains.

Map a repository

tessera repo map --input . --output ./out/repo_pack

Artifacts written:

files.jsonl              one RepoFile per file (path, language, kind, loc, bytes)
repo_map.json            aggregate map (languages, top-level layout, manifests, signals)
index.md                 human-readable map
validation_report.md     hygiene findings
coverage_report.md       files by kind + test-to-source ratio
dependencies_report.md   declared dependencies across all manifests

What it detects

• Languages by extension (Python, JS/TS, Go, Rust, Java, ... plus config/docs/data).
• File kinds: source, test, config, docs, build, data, asset, other.
• Dependency manifests: pyproject.toml, package.json, requirements.txt, Cargo.toml, go.mod — parsed best-effort for declared dependency names.
• Hygiene signals: README, LICENSE, tests, CI config, .gitignore present.

Validation rules

• empty_repo — nothing found after ignoring build/vendor dirs
• missing_readme, missing_license
• no_tests_detected
• no_dependency_manifest, no_ci_config
• large_source_file — a source file over 800 lines of code