This directory contains compact static fixtures for every supported language. The registry covers 60 language families; YAML, JSON, TOML, XML, and Markdown are auxiliary formats exercised through the analyzer auxiliary output.
They are analyzer fixtures, not runnable projects. They intentionally avoid dependency installs, build outputs, vendored code, and generated files.
The per-language directories are the release-grade verification backbone for:
agentskill analyze <repo>aggregate analyzer output- individual analyzer behavior for
scan,measure,config,graph,symbols, andtests - the supported-language matrix advertised by the packaged tool
The mixed/ fixture validates multi-language behavior across several analyzers
in one repository shape.
The extended/ fixture validates the complete expanded registry, including
primary language kinds and the auxiliary format bucket.
Available example repos:
python/javascript/typescript/go/rust/java/kotlin/csharp/c/cpp/ruby/php/swift/objectivec/bash/extended/mixed/
agentskill analyze agentskill-skill/examples/python --pretty
agentskill scan agentskill-skill/examples/typescript --pretty
agentskill evidence agentskill-skill/examples/mixed --prettyUse the installed agentskill CLI as the canonical interface. The example
repos exist to validate analyzers and generation flows; they are not intended
to be built or installed as standalone projects.
SINGLE_LANGUAGE.md, MULTI_LANGUAGE.md, and MONOREPO.md are reference
context examples for the LLM skill. They are useful for designing rich
repository guidance, but they are not the analyzer-fixture source of truth for
language support.