More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Dependency Upgrade Patch Planner

    1

    Turn risky dependency upgrades into an ordered patch plan that maps affected code, separates mechanical edits from semantic migrations, and validates each step with the repository’s own build and test tooling.

    $12

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Dependency Upgrade Patch Planner

    Dependency Upgrade Patch Planner

    Example session with this skill installed

    Upgrade this repository's zod dependency from the current 3.x version to the latest approved 4.x version.

    Repository root
    ./

    Requirements

    • detect the current package manager, manifests, workspaces, and lockfiles;
    • identify every import, wrapper, schema helper, test, and config file affected by the upgrade;
    • use official migration/changelog material when available;
    • separate proposed changes into MECHANICAL_SAFE, MECHANICAL_REVIEW, SEMANTIC_REVIEW, and BLOCKED;
    • do not change business behavior just to satisfy TypeScript;
    • do not replace migration errors with any;
    • preserve the existing package manager;
    • update only dependencies that are actually required for this migration;
    • group proposed changes into an ordered patch plan;
    • run install, typecheck, focused tests, full tests, and build when those commands exist;
    • capture all failures and update the patch plan from the results;
    • generate proposed diffs, an incompatibility matrix, and a final validation report.

    Do not merge or commit changes automatically.

    • Read your context and instructions
    • Compiled the dependency upgrade patch
    • Generated the data export

    Dependency upgrade plan complete

    Target: zod 3.x → 4.x
    Package manager: pnpm
    Workspace packages scanned: 11
    Manifests found: 12
    Lockfiles found: 1
    Files referencing zod: 47

    Breaking-surface summary

    MECHANICAL_SAFE: 21
    MECHANICAL_REVIEW: 9
    SEMANTIC_REVIEW: 6
    BLOCKED: 1

    Highest-impact surface

    File
    packages/api/src/lib/validation.ts

    Risk
    SEMANTIC_REVIEW

    The wrapper centralizes error formatting for 31 schemas.

    The target major version changes behavior used by this helper, so this cannot be treated as a simple import replacement.

    Proposed action
    Update the wrapper first, then validate all downstream schema tests before changing individual callers.

    Patch Group M1 — Manifest + mechanical imports

    Risk
    MECHANICAL_SAFE

    Changes

    • update zod dependency range;
    • refresh pnpm lockfile;
    • apply documented import/API replacements.

    Validation

    pnpm install
    pnpm typecheck
    

    Patch Group S1 — Shared validation wrapper

    Risk
    SEMANTIC_REVIEW

    Changes

    • adapt central validation/error handling;
    • preserve the existing API exposed to application code where practical.

    Validation
    pnpm --filter api test validation
    pnpm --filter api typecheck

    Patch Group M2 — Compiler-driven fixes

    Risk
    MECHANICAL_REVIEW

    Changes

    • update type-level call sites revealed by the compiler;
    • preserve runtime semantics.

    Validation
    pnpm typecheck

    Patch Group T1 — Test migration

    Risk
    MECHANICAL_REVIEW / SEMANTIC_REVIEW

    Changes

    • update assertions only where changed behavior is intentional;
    • leave ambiguous behavioral expectations in the review queue.

    Validation
    pnpm test

    Validation results

    pnpm install: PASS
    pnpm typecheck: PASS after M1 + S1 + M2
    focused API tests: PASS
    full test suite: 1 REVIEW
    production build: PASS

    Remaining review item

    packages/api/test/customer-import.test.ts

    The test depends on legacy validation-error formatting.

    The upgrade can compile and build successfully, but changing this assertion requires confirmation that the new error payload is acceptable to API consumers.

    Status
    SEMANTIC_REVIEW

    Files generated

    dependency_inventory.json
    upgrade_target.json
    affected_references.csv
    breaking_surface_matrix.csv
    patch_plan.md
    proposed_patches/
    validation_report.md
    upgrade_summary.json

    This upgrade plan separates mechanically safe edits from semantic migrations. Review semantic changes and confirm all project tests before merging the dependency upgrade.

    affected_references.csv

    CSV · data export

    Generated

    Example file from a real run - the skill writes it into your workspace.

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

    About this skill

    The problem

    Major dependency upgrades rarely fail in one place.

    Breaking changes can be scattered across imports, configuration, generated types, plugins, tests, and runtime behavior—turning a simple version bump into hours of update → compile → fix → repeat.

    What it does

    Dependency Upgrade Patch Planner turns that migration into a structured sequence of evidence-backed patch groups.

    • Detects manifests, lockfiles, package managers, and workspaces
    • Identifies repository references to the target dependency
    • Maps imports, configs, plugins, wrappers, tests, and generated code
    • Separates mechanically safe edits from semantic migrations
    • Produces an incompatibility matrix
    • Builds an ordered patch plan
    • Runs the repository’s compiler, typecheck, tests, and build commands
    • Uses failures from each iteration to refine the next patch group
    • Preserves the existing package manager and lockfile strategy
    • Records blockers instead of hiding them

    Why this beats prompting it yourself

    A generic prompt can summarize a changelog.

    This skill correlates the upgrade with the actual repository:

    Manifest → Lockfile → Imports → Config → Breaking Surfaces → Patch Groups → Compiler → Tests → Validation Report

    Mechanical vs semantic changes

    The skill explicitly separates edits that can be applied mechanically from changes that require a developer to make a behavior-level decision.

    Repository-safe by design

    The skill does not disable tests, weaken compiler strictness, replace migration errors with any, or upgrade unrelated dependencies just to produce a green build.

    No support for Python or Go. Requires a working local build environment and pre-existing tests to provide validation feedback.

    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 1 month ago

    Requires

    Python

    Permissions

    Terminal / Shell
    Listed1 month ago
    Updated23 days ago

    What's inside

    Frequently Asked Questions

    The skill maps how a specific dependency upgrade impacts your entire codebase, separating simple mechanical updates from complex semantic changes to create a step-by-step, validated patch plan.

    It is compatible with major package managers and supports languages like TypeScript, Java, and Node.js, integrating directly with your existing compiler and test suites for validation.

    No, the skill is designed to respect your repository's integrity; it preserves your lockfile strategy and will never weaken compiler rules or bypass failing tests just to force a build.

    You receive the full skill logic ready for your AI agent, including the multi-step workflow for manifest analysis, incompatibility mapping, and iterative validation reporting.

    Simply point the agent to your repository root and specify the target dependency; the skill automatically detects your manifests, workspaces, and build tools to begin the analysis.

    New in Code Quality & Review