Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Scan project dependencies for vulnerabilities, license issues, and upgrade opportunities across Python, Node.js, Go, and Rust. Use when auditing dependencies, checking licenses, planning upgrades, or assessing supply chain security.
.claude/skills/borghei-dependency-auditor/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-21 | ✗→✓ | ▲ Improved | 75% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 72% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 43% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 52% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 5% | 0% |
A multi-language toolkit for analyzing, auditing, and managing dependencies. It scans manifests and lockfiles across 8+ ecosystems to surface vulnerabilities, classify licenses, detect bloat, and produce safe, phased upgrade plans — giving teams visibility into security, legal, and maintenance risk hidden in their dependency trees.
Before running the audit, confirm these inputs. If any is unknown or vague, ASK — do not assume:
dep_scanner.py vs license_checker.py vs upgrade_planner.py)--policy, --fail-on-high, and the CI verdict)Stop rule: ask only the 2-3 that most change the output. If the user says "just draft it," proceed and list your assumptions at the top of the artifact.
| Tool | Purpose | Command | |------|---------|---------| | dep_scanner.py | Scan manifests/lockfiles across 8+ ecosystems, match CVEs, produce a security report | python scripts/dep_scanner.py /path/to/project --format json --fail-on-high | | license_checker.py | Classify dependency licenses by risk tier and detect conflicts vs the project license | python scripts/license_checker.py /path/to/project --policy strict --warn-conflicts | | upgrade_planner.py | Evaluate semver gaps, assess breaking-change risk, output a phased upgrade plan | python scripts/upgrade_planner.py deps.json --risk-threshold medium --timeline 30 |
Load the reference that matches the task — keep this file lean and pull detail on demand:
This skill covers:
This skill does NOT cover:
| Skill | Integration | Data Flow | |-------|-------------|-----------| | skill-security-auditor | Feed vulnerability scan results into broader security audit workflows | dep_scanner.py --format json output consumed as evidence artifacts | | ci-cd-pipeline-builder | Embed dependency gates in CI/CD pipelines | dep_scanner.py --fail-on-high and license_checker.py --policy strict as pipeline steps | | release-manager | Attach dependency audit reports to release checklists | JSON reports from all three tools included in release documentation | | pr-review-expert | Flag dependency changes during pull request review | Scanner diff between base and head branch dependency files | | env-secrets-manager | Ensure dependency tooling credentials (registry tokens) are securely managed | Registry authentication tokens stored and rotated via secrets manager | | observability-designer | Monitor dependency health metrics over time | Scan summary statistics exported to monitoring dashboards |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-21 | fail→pass | 13,698 | 13,236 | -3% | 1 | 1 | 0% | 2,015 | 3,524 | +75% | 0 | 0 | — |
case-13 | fail→pass | 13,776 | 12,783 | -7% | 1 | 1 | 0% | 2,198 | 3,789 | +72% | 0 | 0 | — |
case-01 | fail→fail | 13,955 | 16,785 | +20% | 1 | 1 | 0% | 980 | 4,425 | +352% | 0 | 0 | — |
case-02 | fail→fail | 9,100 | 14,176 | +56% | 1 | 1 | 0% | 1,671 | 3,924 | +135% | 0 | 0 | — |
case-19 | pass→pass | 14,245 | 11,895 | -16% | 1 | 1 | 0% | 2,315 | 3,562 | +54% | 0 | 0 | — |
case-03 | fail→fail | 16,741 | 6,514 | -61% | 1 | 1 | 0% | 2,929 | 2,637 | -10% | 0 | 0 | — |
case-04 | fail→pass | 13,923 | 12,625 | -9% | 1 | 1 | 0% | 2,678 | 3,824 | +43% | 0 | 0 | — |
case-05 | fail→pass | 8,168 | 4,647 | -43% | 1 | 1 | 0% | 1,531 | 2,324 | +52% | 0 | 0 | — |
case-06 | fail→fail | 18,180 | 11,215 | -38% | 1 | 1 | 0% | 3,393 | 3,685 | +9% | 0 | 0 | — |
case-07 | fail→pass | 12,302 | 5,914 | -52% | 1 | 1 | 0% | 2,231 | 2,336 | +5% | 0 | 0 | — |
case-20 | pass→pass | 11,607 | 6,306 | -46% | 1 | 1 | 0% | 1,751 | 2,441 | +39% | 0 | 0 | — |
case-08 | fail→pass | 6,792 | 7,811 | +15% | 1 | 1 | 0% | 1,309 | 1,909 | +46% | 0 | 0 | — |
case-09 | fail→pass | 7,696 | 3,104 | -60% | 1 | 1 | 0% | 1,476 | 1,938 | +31% | 0 | 0 | — |
case-10 | fail→pass | 13,456 | 3,165 | -76% | 1 | 1 | 0% | 2,335 | 1,977 | -15% | 0 | 0 | — |
case-11 | fail→pass | 4,730 | 4,218 | -11% | 1 | 1 | 0% | 970 | 2,220 | +129% | 0 | 0 | — |
case-12 | pass→pass | 15,759 | 24,802 | +57% | 1 | 1 | 0% | 3,063 | 6,210 | +103% | 0 | 0 | — |
case-14 | pass→pass | 8,543 | 4,617 | -46% | 1 | 1 | 0% | 1,517 | 2,210 | +46% | 0 | 0 | — |
case-15 | pass→pass | 10,810 | 7,611 | -30% | 1 | 1 | 0% | 1,666 | 2,578 | +55% | 0 | 0 | — |
case-16 | pass→pass | 9,677 | 4,268 | -56% | 1 | 1 | 0% | 1,460 | 2,193 | +50% | 0 | 0 | — |
case-17 | pass→pass | 14,766 | 12,184 | -17% | 1 | 1 | 0% | 2,393 | 3,401 | +42% | 0 | 0 | — |
case-18 | fail→pass | 17,794 | 14,203 | -20% | 1 | 1 | 0% | 2,844 | 3,923 | +38% | 0 | 0 | — |
case-22 | fail→pass | 9,251 | 2,447 | -74% | 1 | 1 | 0% | 1,808 | 1,856 | +3% | 0 | 0 | — |
case-23 | fail→pass | 13,762 | 4,055 | -71% | 1 | 1 | 0% | 2,348 | 1,894 | -19% | 0 | 0 | — |
case-24 | fail→pass | 6,813 | 2,508 | -63% | 1 | 1 | 0% | 1,076 | 1,933 | +80% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 24 cases were attempted. The headline lift of +54 percentage points is the difference between those two pass rates over the 24 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.