agentskill@0.9.0 | 2026-04-29

Added

• scripts/lib/generate_runner.py — generate command to create AGENTS.md from repository analysis
• scripts/lib/interactive_runner.py — interactive mode for AGENTS.md generation to prompt for missing inputs
• scripts/lib/reference_flow.py — support for multiple reference repositories in analyze and generate commands
• scripts/lib/output_schema.py — output schema validation with JSON Schema definitions
• scripts/lib/runner.py — enhanced runner with generate command support
• scripts/lib/output.py — output validation improvements
• cli.py — generate subcommand with --interactive flag
• Contract test fixtures for Python and mixed-language examples
• tests/contract_utils.py — shared contract test helpers
• tests/contracts/ — JSON contract files for analyze, config, graph, scan, and symbols
• tests/test_generate_cli.py — generate command CLI tests
• tests/test_interactive_flow.py — interactive flow tests
• tests/test_output_schema.py — output schema validation tests
• tests/test_output_contracts.py — output contract tests
• tests/test_output.py — expanded output validation tests
• AGENTS.md and SKILL.md updated to clarify usage of examples from ClawHub

Changed

• cli.py now exposes generate command alongside analyze and update
• scripts/lib/runner.py updated to support generate command dispatch
• scripts/lib/output.py enhanced with schema-based validation