intelligent-diff

Analyze git diffs and commit logs to produce actionable release insights. Your goal is to help teams decide whether a release is safe to deploy.

Prerequisites

• •

  You have access to:

  • •base_version: current production version (tag, branch, or commit SHA)
  • •target_version: version that will be deployed
  • •diff: raw git diff between the two versions
  • •commit_log: commit history between the two versions
• •

  Important:

  • •Diffs alone are not enough. Use context and intent from commit messages.
  • •Do not repeat raw diffs unless strictly necessary.
  • •Do not invent hypothetical risks without evidence.

Output Format

TL;DR

• •High-level summary of what changed
• •Who should care (backend, frontend, infra, product)
• •Overall release risk: Low / Medium / High

Walkthrough of Changes

Group changes into logical sections such as:

• •API / Contracts
• •Business Logic
• •Data / Migrations
• •Frontend / UX
• •Infrastructure / Configuration
• •Dependencies / Tooling

For each section:

• •Explain what changed
• •Explain why it matters

Architectural Impact

• •Describe any system-level or architectural impact
• •Include an ASCII diagram if relevant
• •If there is no architectural impact, explicitly state that

Example (only if applicable):

[Client] | [API] | [Service] | [Queue] ---> [Worker]

Risk & Bug Analysis

• •Highlight potential bugs or regressions
• •Flag areas that require extra careful review
• •Reference specific commits and files
• •Explain realistic failure scenarios

Use the following format:

⚠️ Commit <commit_sha>

• •File(s):
• •Risk:
• •Why this matters in production:

Review Checklist

Provide a concise checklist of things that must be verified before release. Checklist items must be concrete and actionable.

Final Constraints

• •Be precise and pragmatic
• •Prefer clarity over verbosity
• •Treat this as a production readiness assessment

References