Works with the AI Agents you already use
dependency blast radius auditor
Audit Node.js and Python dependencies for breaking change risks and generate phased upgrade plans.
$7.99
/moSubscription · cancel anytime
Secure checkout via Stripe
- Always the latest version
- Delivered live to your agent via MCP
- Cancel anytime, access ends at period end
dependency blast radius auditor
Example session with this skill installed
Audit this package.json for risks: {"dependencies":{"express":"^4.17.1","axios":"0.21.1","moment":"^2.29.4","lodash":"*"}}
- Read your context and instructions
- Compiled the dependency blast radius
Audit complete for NPM_NODE ecosystem. Detected 4 dependencies with a critical risk profile.
Dependency Blast Radius Audit
| Package | Current | Risk Factors | Score Contribution |
|---|---|---|---|
| moment | ^2.29.4 | Deprecated / EOL | +30 |
| axios | 0.21.1 | Vulnerable Legacy / Pre-1.0 | +37 |
| lodash | * | Unpinned Wildcard | +20 |
| express | ^4.17.1 | Core Framework Pin | +10 |
Total Blast Radius Score: 97 / 100
Verdict: BLOCKED_BREAKING_BLAST_RADIUS
Phased Upgrade Plan
Phase 1: Security & Pinning (Immediate)
# Pin lodash to current stable and patch axios
npm install lodash@4.17.21 axios@0.21.4 --save-exact
Phase 2: Deprecation Migration (Staging)
# Moment is EOL. Migrate to date-fns or Day.js before updating.
# Temporary patch update:
npm install moment@2.30.1
Phase 3: Core Framework
npm install express@4.19.2
Rollback Command
# Deterministic rollback to previous lockfile state
npm ci
Next steps
- Replace
momentwith a modern alternative likeluxonordate-fns. - Replace the
*wildcard forlodashwith a specific version in your manifest. - Run
npm auditto confirm specific CVE patches for theaxiosupgrade.
Subscription · always the latest version
- Secure checkout via Stripe
- Cancel anytime
- Delivered live via MCP
About this skill
Blindly running updates on production manifests often leads to breaking changes and dependency hell. This skill audits your package manifests to quantify upgrade risk and generate safe migration paths.
What it does
- Blast radius scoring calculates a risk score from 0 to 100 based on SemVer stability, deprecation status, and version pinning.
- Manifest parsing extracts dependency trees from Node.js
package.jsonor Pythonrequirements.txtfiles. - Risk classification assigns gate verdicts like
SAFE_AUTO_MERGEorBLOCKED_BREAKING_BLAST_RADIUSto your upgrade path. - Legacy detection flags EOL packages such as
moment,request, ortslintthat require immediate replacement strategy. - Command generation emits exact, phased CLI commands for both the upgrade process and emergency rollbacks.
How it works
- Submit manifest by providing the raw text of your
package.jsonorrequirements.txtfile. - Select policy such as
CONSERVATIVE_PATCH_MINORto limit changes to non-breaking versions orSECURITY_CRITICAL_ONLY. - Analyze report to review the weighted risk sum and identify high-risk unpinned wildcards or unstable pre-1.0 APIs.
- Execute plan using the provided 3-phase sequence for staging and production deployments.
Frameworks & tools
Supports Node.js (npm) and Python (pip) ecosystems. It interfaces with standard manifests like package.json and requirements.txt.
Why this beats prompting it yourself
Generic LLMs often miss specific EOL dates or the nuance of how 0.x.y SemVer ranges differ from stable releases. This skill uses a deterministic scoring formula to ensure risk assessment is consistent across every audit.
Use cases
- Audit legacy repositories to identify deprecated packages before starting a migration.
- Generate rollback commands for CI/CD pipelines to ensure quick recovery from failed updates.
- Sanitize manifests by identifying unpinned
*orlatestversions that cause non-deterministic builds.
Known limitations
Does not execute the actual npm install or pip install commands. It does not perform deep static analysis of code usage, only manifest-level risk.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for LLM Ops.
Does not execute shell commands or install packages. Restricted to npm and pip ecosystems. Does not perform code-level impact analysis.
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 buying a skill.
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
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 1 day ago