Detect deceptive code patterns in AI-generated codebases by analyzing changed files.
npx skills add https://github.com/stronghuni/FindLie --skill find-lie
Please help me install this Agent Skill. Skill: find-lie Source: https://github.com/stronghuni/FindLie/tree/main/find-lie Command: npx skills add https://github.com/stronghuni/FindLie --skill find-lie
Detect lies in AI-generated codebases — including mock data disguised as real data, incomplete implementations, deceptive returns, intent mismatches, disconnected integrations, fake tests, duplicate code, redundant files, and dead code. This Skill empowers engineers to audit generated code for authenticity and correctness across code generation sessions.
Run the FindLie scan on your repository to detect and surface lie code in AI-generated code.
High-intent search queries and answers about installing and using this skill.
Static analysis tools detect mock data and deceptive returns in AI-generated code by cross-checking function names, return values, and unused imports against expected operations to expose intent mismatches and no-ops.
Code forensics for AI-generated codebases is the process of using semantic analysis to audit code authenticity, identifying fake tests, disconnected integrations, and disguised mock data by cross-checking function intent against expected operations.
You audit changed files for fake tests and dead code by running a static analyzer on your current git branch, comparing changed files against a base branch to produce structured findings with evidence and verification commands.
Yes, you can scan for no-op patterns and redundant files outside a git repository, but the analyzer will prompt you to manually choose a scan scope instead of automatically comparing against a base branch.
The best way to find incomplete implementations in agent-generated code is using semantic verification to analyze function intent against expected operations, catching mismatches and dead code through pattern-based lie detection.
Static analysis flags unused imports as deceptive code patterns because they often indicate disconnected integrations or incomplete implementations where AI-generated code fails to execute its stated intent.
anilveersingh1308Review pull request diffs against a base branch for risk and impact.
navapbcExecute tests, lint, type checks, and CI checks before declaring completion.
julianobarbosaReview code for quality, bugs, security issues, and best practices using Codex.
RazisafirAudit a codebase for over-engineering, dead code, and redundant functionality.
