Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Change Risk Review — Contracts, Decisions And Blast Radius

    1

    Review code changes for breaking API contracts, ADR violations, and structural blast radius.

    $12

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Change Risk Review — Contracts, Decisions And Blast Radius

    Change Risk Review — Contracts, Decisions And Blast Radius

    Example session with this skill installed

    Review my staged changes against the main branch. Check for any contract breaks or decision violations in the modules I modified.

    • Read your context and instructions
    • Compiled the change risk review

    I have completed the risk review for the 4 files changed in the working tree compared to main.

    Change Risk Review: Risky

    Executive Summary: This change is classified as

    Risky due to a confirmed public contract break in the core payment module and two untested exports.

    The Change at a Glance

    • Base: main (commit 8f2a1d)
    • Head: Working Tree (staged)
    • Files Touched: src/payments/engine.ts, src/payments/types.ts, src/utils/logger.ts

    Findings

    LabelFindingSource
    ConfirmedRemoved exported function calculateTaxsrc/payments/engine.ts:42
    ConfirmedContradicts ADR-004: "All payment methods must implement Recoverable"docs/adr/004-payment-recovery.md
    LikelyUntested public export validateGatewayResponsesrc/payments/engine.ts:88

    Contract Changes

    The removal of calculateTax is a breaking change. The agent-context explain command reports

    12 consumers across 3 modules (billing-service, invoice-gen, and portal-api) that currently import this symbol.

    Decision and Convention Violations

    • Violation: The new CryptoProvider class does not implement the Recoverable interface.
    • Rule: ADR-004 (Explicit) requires all providers in src/payments/ to support idempotent recovery.

    Blast Radius

    • Primary Module: src/payments/engine.ts
    • Total Consumers: 18 files.
    • Risk: High. Changing the internal state machine here affects the entire checkout flow.

    Next steps

    • Restore the calculateTax export or provide a deprecated alias to avoid breaking the billing-service.
    • Update CryptoProvider to implement the Recoverable interface as per ADR-004.
    • Add unit tests for the newly exported validateGatewayResponse function.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Standard code reviews catch bugs and style issues, but they often miss systemic risks like broken public contracts, violated architectural decisions, or hidden blast radii. This skill performs a deep-impact analysis on changes already made to a repository to identify the failures that pass every linter yet break production.

    What it does

    • Public contract validation checks if the change removes or renames symbols that external consumers depend on.
    • Decision enforcement verifies the diff against established Architecture Decision Records (ADRs) to flag rule violations.
    • Blast radius mapping calculates exactly how many files and consumers are affected by changes to a specific module.
    • Structural exposure analysis identifies if the change introduces layering violations or circular dependencies.
    • Test coverage auditing flags new public exports that lack corresponding test files.

    How it works

    1. Frame the review by identifying the working tree, specific commit, or branch to be analyzed.
    2. Collect context using git and the optional ai-agent-context CLI to build a dependency graph and load team conventions.
    3. Execute checks against the diff to find contract breaks, ADR contradictions, and structural risks.
    4. Deliver a verdict of Safe, Safe with Caveats, or Risky based on the evidence found.

    Frameworks & tools

    The skill primarily utilizes git for diff analysis and file inspection. It enhances results using the ai-agent-context CLI for structural mapping and decision tracking.

    Why this beats prompting it yourself

    Generic prompts miss the "invisible" dependencies and team-specific rules stored in ADRs. This skill enforces a rigorous six-check framework that forces the agent to prove every finding with file-level evidence and consumer counts.

    Use cases

    • Reviewing a pull request for breaking API changes before a release.
    • Checking a branch against existing ADRs to ensure architectural compliance.
    • Assessing the merge risk of a large refactor by mapping its full blast radius.
    • Auditing the working tree to find untested exports before staging.

    Known limitations

    The skill is read-only and will not fix the code it reviews. Structural exposure analysis (Check 4) requires the ai-agent-context CLI and is unavailable in the git-only fallback mode.

    Read-only; does not fix code. Structural checks require ai-agent-context CLI. Fallback mode provides limited consumer counts. Use for post-change review only, not pre-change planning.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 28 days ago

    Requires

    ai-agent-context
    Node.js
    agent-context
    Git

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access
    Listed28 days ago

    What's inside

    Frequently Asked Questions

    Unlike standard linters that check syntax, this skill performs an impact analysis to detect breaking API changes, violations of your team's Architecture Decision Records (ADRs), and the 'blast radius' of structural changes that could cause downstream failures.

    The skill requires `git` for basic diff analysis. To unlock the full potential of blast radius mapping and structural exposure analysis, you will need to install the `ai-agent-context` CLI in your environment.

    Yes, it is designed to be framework-agnostic. While it uses `git` for analysis, it can review code across various languages as long as the project follows standard architectural patterns or maintains ADRs.

    This skill is built for agents that have filesystem and terminal access, such as those running in VS Code environments or local CLI-based agents that can execute git commands.

    The skill is read-only and functions as a diagnostic auditor; it will identify and explain risks like circular dependencies or contract breaks, but it will not modify your source code to fix them.

    You receive the skill definition, a specialized six-check analysis framework, and documentation on how to integrate your team's specific ADRs into the review process.

    New in Code Quality & Review