Codebase Impact Analyst — Map Changes Before You Code

Identifying the blast radius of a code change manually is error-prone and slow. This skill automates codebase impact analysis, ensuring every dependency, consumer, and integration point is accounted for before a single line of code is written. It provides a structured, evidence-based report to help developers plan features, refactors, and bug fixes with high confidence.

What it does

• Blast radius mapping identifies every module, file, and symbol affected by a proposed change.
• Dependency discovery finds callers and consumers across the entire repository to prevent regressions.
• Risk assessment categorizes potential breakage in public APIs, data schemas, and shared modules.
• Validation planning generates a targeted list of tests to run and manual verification steps to take.
• Diff auditing analyzes existing git diffs to assess architectural impact and compliance.

How it works

1. Analyze environment by checking repository root, stack configuration, and the presence of the ai-agent-context CLI.
2. Scan codebase using specialized analysis modes for proposed features, existing diffs, or module deep-dives.
3. Verify findings by cross-referencing tool output with actual source code to confirm structural relationships.
4. Generate report using a standardized template that includes an impact table, risk map, and implementation sequence.

Frameworks & tools

Works with any git-based repository. It has optimized support for Node.js, TypeScript, Python, Go, Rust, and Java environments. It uses git and the optional ai-agent-context CLI for deterministic dependency graph analysis.

Why this beats prompting it yourself

Generic LLM prompts often hallucinate imports or miss dynamic dependencies in large repos. This skill follows a strict evidence-based workflow that requires confirmation from source files and specialized CLI tools, ensuring the "unknowns" are explicitly identified rather than ignored.

Use cases

• Feature planning to map out exactly which files need modification for a new requirement.
• Refactoring safety to ensure changing a shared utility won't break downstream consumers.
• Code review to audit the architectural impact of a complex pull request before merging.
• Regression prevention by identifying the minimum set of tests required to validate a specific change.

Known limitations

Does not perform code implementation. Dynamic imports, reflection, and network-based calls may require manual verification. Analysis depth is limited by the local environment configuration.