dependency intelligence upgrade manager
Audit dependencies, triage security vulnerabilities, and generate safe, batched upgrade plans for npm, Python, and Go.
$7
Works with the AI Agents you already use
dependency intelligence upgrade manager
Example session with this skill installed
Run a dependency inventory on this project and triage the attached osv-scanner.json output. Check for high-risk breaking changes in the proposed upgrades.
- Read your context and instructions
- Compiled the dependency intelligence upgrade
I have completed the inventory and triaged the vulnerabilities for the project. The inventory covered 142 packages across npm and Python ecosystems with 98% coverage.
Dependency Audit & Upgrade Plan
| Package | Current | Target | Priority | Risk Level | Issue |
|---|---|---|---|---|---|
lodash | 4.17.15 | 4.17.21 | P1 | CHG-LOW | CVE-2020-8203 (ReDoS) |
requests | 2.22.0 | 2.31.0 | P1 | CHG-MEDIUM | CVE-2023-32681 (Leak) |
fastapi | 0.68.0 | 0.100.0 | P2 | CHG-HIGH | Breaking changes in middleware |
django | 3.2.0 | 3.2.20 | P1 | CHG-LOW | CVE-2023-36053 |
Breaking Changes Analysis
fastapi (0.68.0 -> 0.100.0): Detected
CHG-HIGH. Major version jump. Starlette internal changes may break custom exception handlers. Usage of TestClient needs verification against new httpx requirements.
requests (2.22.0 -> 2.31.0): Detected
CHG-MEDIUM. Minor version change includes security fixes for proxy handling.
Upgrade Plan (Batch 1: Security Priority)
- Command:
npm install lodash@4.17.21- Validation: Run
npm test - Rollback:
npm install lodash@4.17.15
- Validation: Run
- Command:
pip install requests==2.31.0 django==3.2.20- Validation: Run
pytest - Rollback:
pip install requests==2.22.0 django==3.2.0
- Validation: Run
Next steps
- Review the
fastapimigration guide before proceeding with Batch 2. - Execute Batch 1 commands in a dedicated branch.
- Verify that the
LIC-REVIEWflag on the newfastapitransitive dependencies meets company policy.
Connects securely to your tools. The creator never sees your data.
About this skill
Struggling to prioritize a mountain of Dependabot alerts or debating if a major version bump will break your production environment? Managing dependencies is often a manual slog of reading changelogs and guessing at reachability.
This skill automates the analysis of package manifests and lockfiles to turn raw vulnerability data into a validated upgrade strategy. It bridges the gap between seeing a CVE and knowing if your project is actually at risk, focusing on reachability, SemVer impact, and license compliance.
What it does
- Multi-ecosystem inventorying processes npm, Python, and Go manifests to map direct and transitive dependencies.
- Vulnerability prioritization triages advisories (OSV, CVE, KEV) into P1-P3 levels based on real risk rather than just generic scores.
- Breaking change analysis evaluates SemVer transitions and release notes to flag risky 0.x updates or major breaking changes.
- License policy enforcement audits SPDX identifiers against your specific allow/deny lists to prevent legal drift.
- Batch upgrade planning generates ordered, testable sequences for updates with defined rollback commands.
How it works
- Inventory phase uses provided scripts to scan
package-lock.json,requirements.txt, orgo.modfor a full dependency graph. - Triaging imports external audit data (like
npm auditorosv-scanner) to match vulnerabilities against the current inventory. - Evaluation reviews changelogs and SemVer deltas to classify risks as high, medium, or low impact.
- Planning generates a structured upgrade plan that groups security fixes first and isolates major version jumps.
Frameworks & tools
Built for projects using npm, pip, or go. It utilizes OSV.dev data, SPDX license standards, and signals like EPSS and KEV for risk assessment.
Why this beats prompting it yourself
Generic prompts often hallucinate version numbers or fail to account for transitive dependency depth. This skill uses deterministic scripts to ensure every version, CVE ID, and license claim is sourced from your actual files or official databases.
Use cases
- Auditing a legacy codebase to identify abandoned or high-risk dependencies.
- Triage of high-volume Dependabot or Renovate alerts to find what actually needs fixing today.
- Preparing a secure SBOM (Software Bill of Materials) for compliance or procurement.
- Evaluating the safety of a major framework migration.
Known limitations
Does not automatically execute npm install or pip install commands. Requires external advisory data (JSON) to be provided for vulnerability matching.
Requires providing advisory JSON files for triage. Does not perform automated code changes or package installation without manual approval. Limited to npm, Python, and Go ecosystems.
How to use
- 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
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 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 3 days ago
Requires
Permissions
Allowed Hosts