
dev-dependency-management
Guides dependency audits and upgrades. Use when auditing dependencies, upgrading vulnerable packages, managing lockfiles or package managers, SBOMs, or update and monorepo policy.
- Installs
- 172
- Last 24h
- —
- Trending rank
- —
- Security
- Not audited
Install
Works with every agent the skills CLI supports.
npx skills add https://github.com/vasilyu1983/ai-agents-public --skill dev-dependency-management- 01
Install
Run the command in your project. The CLI asks which agents to install for.
- 02
Describe the task
Ask your agent for something this skill covers. It loads dev-dependency-management automatically when relevant.
- 03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.
SKILL.md
The instructions your agent reads when the skill is triggered.
Dependency Management
Quick Reference
| Task | Use |
|---|---|
| Ecosystem defaults and package-manager choice | references/ecosystem-guides.md |
| Lockfiles and CI install policy | references/lockfile-management.md |
| Security scanning, SBOMs, and provenance | references/security-scanning.md, assets/automation/template-supply-chain-security.md, assets/automation/template-sbom-vuln-triage-checklist.md |
| Monorepos and workspace policy | references/monorepo-patterns.md, assets/nodejs/pnpm-workspace-template.yaml |
| Update strategy and rollback | references/update-strategies.md, assets/automation/template-dependency-upgrade-playbook.md |
| Add-or-avoid dependency decision | references/dependency-selection-guide.md, references/transitive-dependencies.md |
| Vulnerability detection | The ecosystem's native audit tool (npm audit, pnpm audit, pip-audit, cargo audit, govulncheck, composer audit) or osv-scanner; see references/security-scanning.md |
| Vulnerability triage, VEX, EU CRA lookup | references/security-scanning.md |
| Policy-manifest scorer (not a vulnerability scanner: no advisory database, refuses real manifests, exits 2 on unchecked input) | python3 scripts/dep_auditor.py --help, limits in scripts/README.md |
Route Elsewhere
- Framework-specific frontend or backend implementation: use the relevant software skill.
- AppSec architecture beyond the dependency layer: use software-security-appsec.
- CI/CD platform design: use ops-devops-platform.
Workflow
- Identify ecosystem, package manager, lockfile, and wrapper conventions already present.
- Decide the smallest safe change: add, remove, pin, update, audit, or migrate.
- Load only the guidance needed for lockfiles, security, monorepo policy, or update strategy.
- Before migrating Node tooling, check the target major's release notes and its script-approval settings using the lookups below; compare them with CI's runtime and required native builds.
- Finish with reproducibility checks, rollback notes, and any audit or SBOM follow-up.
Upgrade proof and rollback bundle
Treat the manifest and lockfile as one change unit. Before the upgrade, capture the resolved version, integrity/source metadata, transitive dependency delta, install-script delta, and the exact build/test/runtime probes that protect the affected path. Review newly introduced packages and lifecycle scripts even when the requested direct dependency looks harmless.
A rollback instruction must restore both manifest and lockfile to a known pair and state whether caches, generated clients, database state, or persisted wire formats make code rollback insufficient. For a stateful or protocol-changing upgrade, require a forward-compatible migration or a tested downgrade path before rollout. “Revert the dependency bump” is not a complete rollback plan when the new version has already written data or changed an external contract.
Core Decisions
Package-Manager Defaults
| Ecosystem | Default for new repos | Version lookup | Key constraint |
|---|---|---|---|
| Node | pnpm unless compat pressure favors npm | npm view pnpm dist-tags (a new major may ship on a non-default tag) | pnpm 11 removed onlyBuiltDependencies in favour of allowBuilds; check the target major's release notes for its runtime and install path, and build settings for script approvals before setting CI policy |
| Python | uv | uv release notes | Check whether uv is still on 0.x versioning before promising API stability |
| Rust | Cargo | pinned rust-toolchain.toml | commit Cargo.lock for apps |
| Go | go modules | go / toolchain lines in go.mod | go.mod + go.sum canonical |
| Java | Maven wrapper or Gradle wrapper | wrapper properties | wrappers plus BOMs or version catalogs |
| .NET | PackageReference | global.json SDK pin | PackageReference over packages.config |
| PHP | Composer | composer.json config.platform | commit composer.lock for apps |
Keep repo-local consistency more important than theoretical ecosystem purity.
Lockfile and Toolchain Policy
Minimum rules:
- commit application lockfiles
- use exact lockfile installs in CI
- avoid hand-editing lockfiles
- do not carry multiple lockfiles for one package graph
- prefer repo-local wrappers or toolchain files over unpinned global installs
Rule: rules/deps/lockfiles.md loads this invariant when Claude edits a matching file.
Update Strategy
Default cadence:
- patch: small and frequent
- minor: batched and tested
- major: isolated with release-note review and rollback plan
- security: prioritize by reachability and exploitation evidence (KEV listing first, EPSS as a likelihood signal, CVSS as severity only), and record
not_affectedfindings as VEX statements with a justification; see Vulnerability Triage
Supply-Chain Controls
Use:
- official registries where possible
- provenance or signature verification where supported
- SBOM generation for release artifacts, with the minimum-element fields your customer or regulator requires (lookup step in references/security-scanning.md)
- explicit review of install or build scripts
- a build-time static gate on every shipped artifact, not only source code. The gate covers the package or binary and everything bundled into it: vendored and statically linked dependencies, plugins, extension or recipe manifests, and templates. Before publish it checks licenses against an allowlist, known advisories, and allowed sources (no git or path dependencies that bypass the lockfile), and it schema-validates bundled manifests. An artifact that fails does not reach a release channel. A policy file such as cargo-deny's
deny.tomlis one way to express this - expiration on overrides, resolutions, and temporary pins
- a release-age delay (pnpm
minimumReleaseAge, RenovateminimumReleaseAge, Dependabotcooldown) to avoid consuming just-published packages; it is one layer, not a fix, because several historical npm compromises were live for only hours before removal (see the incident patterns in references/security-scanning.md) - dependency confusion mitigations: scope all internal packages, audit all org-scoped packages before use
- EU market: if you ship products with digital elements in the EU, look up which Cyber Resilience Act (Regulation (EU) 2024/2847) vulnerability-reporting obligations apply to you before designing the triage workflow; see EU Cyber Resilience Act Lookup
- Before adopting an npm major, read its versioned configuration docs and changelog for dependency lifecycle-script approvals and git/remote dependency defaults. Check required native builds and sources in CI, and review each approval rather than enabling all scripts.
AI-Generated Dependency Risk
Before accepting an AI-suggested package:
- verify it exists
- check for typosquatting risk
- check maintenance and release cadence
- prefer standard library or existing dependencies if feasible
- run the same audit and review workflow as for any other new package
Rule: rules/common/dependencies.md loads this invariant in every coding session.
Output Modes
Default to one of these:
- Dependency policy brief: manager choice, lockfile rules, update cadence, and security controls.
- Upgrade plan: scope, batching, testing, rollback, and audit follow-up.
- Dependency audit: health risks, unmaintained packages, graph complexity, and remediation order.
- Monorepo dependency strategy: workspace model, shared policy, and update automation rules.
Known Traps
-
Upgrading transitive or security-sensitive packages without checking whether the fix actually lands in the production artifact path.
-
Mixing wrapper, lockfile, and package-manager upgrades in one move, which makes rollback and blame assignment much harder.
-
Assuming monorepo hoisting or workspace dedupe is harmless when postinstall scripts, peer deps, or native builds are involved.
-
Accepting temporary pins or overrides without a removal owner, expiry, and retest trigger.
-
Treating SBOM generation as complete supply-chain control while provenance, install scripts, and release process remain unreviewed.
-
Treating a clean
dep_auditor.pyrun as an audit: it reads self-declared flags only. Its input requires an explicit dependency list (use[]for a dependency-free ecosystem), boolean policy flags, non-negative integer ages when provided, and a scan date that is not in the future; malformed declarations exit 2.
References
| File | What it covers |
|---|---|
| references/ecosystem-guides.md | Per-ecosystem package-manager defaults, CI install commands, and watchouts for Node, Python, Rust, Go, Java, .NET, PHP |
| references/lockfile-management.md | Lockfile matrix, golden rules, per-ecosystem exact-install commands, CI rules, and drift recovery |
| references/security-scanning.md | Native audit commands, SBOM generation, provenance controls, Dependabot/Renovate usage, and triage workflow |
| references/monorepo-patterns.md | JS/TS workspace defaults, pnpm supply-chain settings, polyglot structure, and version governance |
| references/dependency-selection-guide.md | Add-or-avoid decision criteria, graph inspection commands, AI-suggested package checklist |
| references/update-strategies.md | Update cadence table, batch-by-risk workflow, bot policy, and rollback rule |
| references/transitive-dependencies.md | Tree inspection, override patterns, deduplication, and resolution decision tree |
| references/license-compliance.md | License risk table, GPL decision tree, automated tooling, CI integration, and SBOM generation commands |
| references/version-conflict-resolution.md | Conflict types, per-manager diagnostic commands, forced resolution syntax, and pnpm catalogs |
| references/container-dependency-patterns.md | Multi-stage build patterns, layer caching, vulnerability scanning (Trivy/Grype), and reproducible base image pinning |
| references/semver-guide.md | SemVer constraint syntax for npm, Python, and Cargo with common pitfalls |
| references/anti-patterns.md | Critical and moderate anti-patterns with corrective examples |
Curated source links live in data/sources.json.
Navigation
- Assets: assets/nodejs/package-json-template.json, assets/nodejs/npmrc-template.txt, assets/nodejs/pnpm-workspace-template.yaml, assets/python/pyproject-toml-template.toml, assets/automation/dependabot-config.yml, assets/automation/renovate-config.json, assets/automation/audit-checklist.md, assets/automation/template-dependency-upgrade-playbook.md, assets/automation/template-supply-chain-security.md, assets/automation/template-sbom-vuln-triage-checklist.md
- Scripts and samples:
scripts/dep_auditor.py,scripts/README.md, data/sample-dependency-manifest.example.json
Learnings Loop
When prior decisions or pitfalls are relevant, consult learnings.consolidated.md if present; use learnings.md only for needed history or as the available fallback. Otherwise skip both.
After applying it, if you encountered a pattern worth remembering, a mistake worth preventing, or a domain fact that surprised you, append one dated bullet to learnings.md via agents-skills-feedback-loop/scripts/append_learning.py. Do not modify SKILL.md itself.
FAQ
What does the dev-dependency-management skill do?
Guides dependency audits and upgrades. Use when auditing dependencies, upgrading vulnerable packages, managing lockfiles or package managers, SBOMs, or update and monorepo policy.
How do I install dev-dependency-management?
Run npx skills add https://github.com/vasilyu1983/ai-agents-public --skill dev-dependency-management in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/dev-dependency-management/ (personal) or .claude/skills/dev-dependency-management/ (project).
Which AI agents can use dev-dependency-management?
Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.
Is dev-dependency-management safe to install?
No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.
Who maintains dev-dependency-management?
It is published by vasilyu1983 in vasilyu1983/ai-agents-public. It has 172 recorded installs.
More from vasilyu1983/ai-agents-public
software-architecture-design
vasilyu1983/ai-agents-public
Designs runtime topology in a chosen solution. Use when choosing monolith vs services, cells, consistency, ADRs. Not for cross-system design (software-solution-architecture).
product-management
vasilyu1983/ai-agents-public
Guides founder-PM discovery, roadmaps, prioritization, and PMF measurement. Use when planning product strategy, metrics, or roadmaps.
document-xlsx
vasilyu1983/ai-agents-public
Builds Excel charts, pivots, validation, exports, Office Scripts/Graph automation, and workbook audits. Use when automating workbooks; for plain edits prefer official xlsx skill.
software-ui-ux-design
vasilyu1983/ai-agents-public
Designs and audits UI/UX systems with usability and accessibility requirements. Use when shaping flows, design systems, interaction patterns, or WCAG-aware product behavior.
document-pdf
vasilyu1983/ai-agents-public
Handles PDF/UA accessibility, real redaction, metadata scrubbing, signing, and extraction. Use when redacting or signing PDFs; for plain create/merge prefer the official pdf skill.
qa-testing-playwright
vasilyu1983/ai-agents-public
Builds and debugs Playwright E2E suites. Use when authoring browser tests, fixing flakes, or hardening Playwright CI and locator strategy.
Popular ai & agents skills
google-agents-cli-publish
google/agents-cli
This skill should be used when the user wants to "publish an agent", "publish my ADK agent", "register an agent with Gemini", "publish to Gemini" (formerly Gemini Enterprise), or needs guidance on the agents-cli publish gemini command. Also use when the user wants to "manage agents in Agent Registry", "list/update/delete registered agents", or "register an MCP server". Covers ADK vs A2A registration modes, programmatic and interactive usage, flag reference, auto-detection from deployment metadata, Agent Registry fleet management, and troubleshooting. Part of the agents-cli skills suite. Do NOT use for deployment (use google-agents-cli-deploy).
google-agents-cli-eval
google/agents-cli
This skill should be used when the user wants to "run an evaluation", "evaluate my agent", "evaluate my ADK agent", "write an eval dataset", "analyze eval failures", "compare eval results", "optimize agent", or needs guidance on the Gemini platform eval methodology and the Quality Flywheel. Covers eval metrics, dataset schema, LLM-as-judge scoring, and common failure causes. Applies to any agents-cli project, whatever framework the agent is written in. Do NOT use for agent API code patterns (ADK: use google-agents-cli-adk-code), deployment (use google-agents-cli-deploy), or project scaffolding (use google-agents-cli-scaffold).
skill-creator
anthropics/skills
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
codex-pet
prime-skills/runcomfy-agent-skills
Codex Pet generator on RunComfy. Build a Codex-compatible Codex Pet spritesheet.webp + pet.json from a single reference image, drop it into ${CODEXHOME:-$HOME/.codex}/pets/ / and Codex picks it up as a custom Codex Pet next to the 8 built-ins. This skill produces the exact Codex Pet atlas Codex expects (1536x1872 PNG/WebP, 8 cols x 9 rows, 192x208 cells, 9 animation states — idle, running-right, running-left, waving, jumping, failed, waiting, running, review). Calls OpenAI GPT Image 2 edit ONCE via the local RunComfy CLI as runcomfy run openai/gpt-image-2/edit to produce a canonical Codex Pet pose, then assembles all 9 animation rows programmatically with ImageMagick micro-transforms — no Codex Pro, no $imagegen, no OPENAIAPIKEY required, only RUNCOMFYTOKEN. Triggers on "codex pet", "create codex pet", "make codex pet", "hatch codex pet", "/hatch image", "desktop pet codex", "codex pets", "spritesheet.webp", or any explicit ask to build a custom pet for OpenAI Codex.
full-output-enforcement
leonxlnx/taste-skill
Overrides default LLM truncation behavior. Enforces complete code generation, bans placeholder patterns, and handles token-limit splits cleanly. Apply to any task requiring exhaustive, unabridged output.
executing-plans
obra/superpowers
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available