review-changes

Reviews code changes using risk-scored change detection and knowledge graph impact analysis.

Updated Aug 4, 2026
One-click install
npx skills add https://github.com/sarthkdobriyal/hisaabi --skill review-changes-sarthkdobriyal
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: review-changes
Source: https://github.com/sarthkdobriyal/hisaabi/tree/main/.gemini/skills/review-changes
Command: npx skills add https://github.com/sarthkdobriyal/hisaabi --skill review-changes-sarthkdobriyal

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Manual code reviews often miss hidden blast radius and untested changes. This Skill performs a structured, risk-aware review by detecting changes, tracing impacted execution flows, and checking test coverage through a knowledge graph. ## Core Features & Use Cases - Risk-Scored Change Detection: Runs detect_changes_tool to surface high, medium, and low risk modifications. - Impact Analysis: Uses get_affected_flows_tool and get_impact_radius_tool to map which execution paths and downstream code are affected. - Test Coverage Checks: Queries the graph for tests covering high-risk functions and suggests specific test cases for untested changes. - Token-Efficient Workflow: Enforces minimal-context retrieval and a budget of 5 tool calls and 800 output tokens per review. - Use Case: Before merging a pull request, run the review to get findings grouped by risk level, coverage status, and a merge recommendation. ## Quick Start Ask the assistant to review the current code changes and produce a risk-grouped report with test coverage status and a merge recommendation.

Frequently Asked Questions about review-changes

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I review code changes with a knowledge graph?▼

Start with get_minimal_context, then run detect_changes_tool for risk-scored changes, get_affected_flows_tool for impacted paths, and get_impact_radius_tool for blast radius. Finish by checking test coverage with query_graph_tool using the tests_for pattern.

How to check test coverage for changed functions?▼

Run query_graph_tool with pattern set to tests_for on each high-risk function identified by the change detection step. For any untested changes, the workflow suggests specific test cases to add.

What output does a risk-based code review produce?▼

Findings are grouped by risk level (high, medium, low) and include what changed and why it matters, test coverage status, suggested improvements, and an overall merge recommendation.

How many tool calls does the review workflow allow?▼

The workflow targets completing any review, debug, or refactor task in at most 5 tool calls and 800 total output tokens. It requires starting with get_minimal_context and using detail_level minimal before escalating to standard.

When should I escalate from minimal to standard detail level?▼

Escalate to standard detail level only when the minimal output is insufficient to answer the review question. The default for all graph calls is minimal to preserve token efficiency.