Detect and fix code-quality issues left behind by AI coding agents.
Deterministic detection for dead code, narrative comments, and hidden fallbacks.
aislop is a deterministic CLI tool that scans codebases for the hidden technical debt generated by AI coding assistants like Claude Code and Cursor. It evaluates code across 10 language targets with 50+ rules without running an LLM at runtime.
It runs a deterministic scan across your repository to output a score from 0 to 100, checking for issues such as swallowed exceptions, unsafe casts, and hallucinated imports. You can use it to auto-fix mechanical problems or feed diagnostic data directly back into your coding agent.
It suits engineering teams adopting AI-generated workflows who need to enforce code hygiene and compliance without relying on runtime LLM calls. It does not suit teams working primarily outside its supported 10 language targets.
Scans across 10 language targets without any LLM in the runtime path.
Supports CLI execution, CI mode with JSON/SARIF output, and GitHub Actions.
Allows ignoring specific lines or files using custom comment directives.
Overrides rule severities or disables specific checks via a project YAML config.
Free under its open-source license, with badges provided via scanaislop.com. The only cost is running the tool yourself in your local environment or CI.
896 open remote AI-training roles, updated November 16, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
No, aislop relies entirely on 50+ deterministic rules for sub-second, consistent code analysis.
If your project is mostly written in unsupported languages, aislop withholds the score and reports a coverage breakdown rather than outputting inaccurate metrics.
Yes, you can use inline comment directives like `aislop-ignore-next-line` with optional rule names to silence specific diagnostics.
See Zekai first in Google