Audit whole-codebase tech-debt with severity-ranked findings.
npx skills add https://github.com/AgentSystemLabs/core --skill audit-agentsystemlabs
Please help me install this Agent Skill. Skill: audit Source: https://github.com/AgentSystemLabs/core/tree/main/plugins/agentsystem-core/skills/audit Command: npx skills add https://github.com/AgentSystemLabs/core --skill audit-agentsystemlabs
It helps you systematically find and remove accumulated tech-debt across an entire codebase, beyond what diff-only reviews catch, so the result is production-ready and easier to maintain.
Use it when you want an end-to-end “deep clean” or a production readiness pass, especially after growth, churn, or unclear ownership that has caused design and quality drift.
Ask your agent: audit the codebase for tech-debt and apply safe mechanical fixes while gating any structural changes.
High-intent search queries and answers about installing and using this skill.
A whole-codebase tech-debt audit maps architecture and data flow, runs repo-wide checks for types, security, and performance, then applies mechanical fixes while gating structural changes. Deduplicated findings are severity-ranked for safe refactoring without behavior changes.
Finding accumulated tech-debt requires mapping entry points and boundaries, then running a fan-out of type hardening, duplication, and security checks across the repo. Diff-only reviews miss systemic drift, so a consolidated audit deduplicates findings into a severity-ranked report.
Yes, production readiness hardening runs a battery of audit checks across code, contracts, concurrency, and observability. Findings are deduplicated and grouped by severity, enabling targeted remediation of data integrity and accessibility issues before deployment.
Yes, baseline gates including typecheck, lint, formatter, and tests are required. Explicit scoping inputs for scope, skip, and mode define boundaries, while interactive confirmation and gated application with re-verification ensure safe mechanical fixes.
Gated remediation applies mechanical fixes by default and routes structural or behavior-adjacent changes through explicit approval. Re-verification runs after application, ensuring refactoring and cleanup maintain baseline integrity across code and architecture.
A deep cleanup runs specialized audit checks covering security hardening, data integrity, concurrency, accessibility, and performance. It maps data flow to ground findings, producing deduplicated results grouped by severity for actionable remediation.
adiomasRun typecheck, lint, tests, and build in deterministic order.
tronghieuRun the Lumina-Wiki linter in JSON mode and classify findings by severity.
quangphu1912Trace data flows through codebases to detect security vulnerabilities and integrity issues.
webdevcodyRun repository-wide tech-debt audits and consolidate severity-ranked findings.
