Skylos - Python SAST & Dead Code Detection
Version updated for https://github.com/duriantaco/skylos to version v3.4.1.
- This action is used across all versions by 11 repositories.
Action Type
This is a Composite action.
Go to the GitHub Marketplace to find the latest changes.
Action Summary
Skylos is a privacy-first static analysis tool for Python, TypeScript, and Go that automates the detection of dead code, security vulnerabilities (e.g., SQL injection, secrets leaks), and code quality issues. It uses a hybrid engine combining AST analysis and optional AI agents to improve precision by eliminating false positives and identifying deep logic errors missed by traditional linters. Skylos integrates seamlessly into CI/CD pipelines, providing automated code audits, inline PR reviews, and quality gates to enhance development workflows.
Release notes
[3.4.1] - 2026-02-21
Added
- BFS from entry points through the import graph to eliminate false positives for code in orphaned modules
__getattr__package handling and dynamic module safety for lazy-loading andeval/execpatterns- Relative import resolution in raw import collection
skylos creditscommand and credit check before--upload- MCP server auth, rate limiting, and per-tool credit deduction
Fixed
--trace --jsonand--pytest-fixtures --jsonproducing invalid JSON (capture_outputlogic was inverted, letting subprocess stdout leak into JSON output)
Changed
- MCP tools now gated with auth check + credit deduction before execution
- Unreachable module definitions capped at confidence 5