dev-toolbelt/dev-team-agents1 installs

spawn-classifier

Q: How do I install this agent skill?

Run npx skills add https://github.com/dev-toolbelt/dev-team-agents --skill spawn-classifier.

Q: Is this agent skill safe to install?

Review the displayed partner audits, complete source, permissions, dependencies, and exact revision before installing.

Conditional agent spawn — path patterns and text triggers.

How do I install this agent skill?

npx skills add https://github.com/dev-toolbelt/dev-team-agents --skill spawn-classifier

view source ↗

Is this agent skill safe to install?

Gen Agent Trust Hubpass
The skill is a classification engine that routes tasks to specific AI agents (e.g., backend, frontend, security) based on file extensions or keywords found in the user's request. It contains only instructional logic and no executable code, network operations, or credential handling.
Socketpass
No alerts
Snykpass
Risk: LOW · No issues

What does this agent skill do?

File-Path Triggers

Evaluate changed files (from git diff --name-only or the task description) against these patterns:

Changed path pattern	Agent to spawn
`*.{go,py,rb,php,java,kt,rs}`	`backend-developer`
`app/Http/`, `app/Services/`, `src/controllers/`, `routes/`	`backend-developer`
`*.{tsx,jsx,vue,svelte}`	`frontend-developer`
`src/components/`, `src/pages/`, `resources/js/`, `app/javascript/`	`frontend-developer`
`database/migrations/`, `db/migrate/`, `prisma/`, `*.sql`	`database-specialist`
`.github/workflows/`, `Dockerfile`, `docker-compose.yml`	`devops-specialist`
`.tf`, `.tfvars`, `Pulumi.yaml`, `serverless.yml`	`devops-specialist`
`Jenkinsfile`, `.gitlab-ci.yml`, `*.circleci/`	`devops-specialist`
`.spec.{ts,tsx,js}`, `.test.{ts,tsx,js}`, `tests/`, `spec/` (frontend)	`frontend-test-specialist`
`*.spec.{py,rb,go,java,kt}`, `tests/`, `spec/` (backend)	`backend-test-specialist`
`*.stories.{ts,tsx,js}`, `storybook/`, `.storybook/`	`frontend-developer` + `ui-ux-designer`

Text / Intent Triggers

Evaluate the task description or user message:

Keyword or phrase	Agent to spawn
"security", "vulnerability", "CVE", "auth bypass", "injection", "XSS", "CSRF"	`security-specialist`
"penetration test", "threat model", "OWASP"	`security-specialist` + `software-architect`
"test", "spec", "coverage", "unit test", "integration test" (frontend context)	`frontend-test-specialist`
"test", "spec", "coverage", "unit test", "integration test" (backend context)	`backend-test-specialist`
"design", "UX", "wireframe", "mockup", "user flow", "component library"	`ui-ux-designer`
"design system", "token", "Figma"	`ui-ux-designer` + `frontend-developer`
"database", "schema", "migration", "index", "query", "ORM"	`database-specialist`
"deploy", "CI/CD", "pipeline", "Docker", "Kubernetes", "infra"	`devops-specialist`
"architecture", "ADR", "design decision", "trade-off", "system design"	`software-architect`
"API", "REST", "GraphQL", "endpoint", "service"	`backend-developer`
"bug", "fix", "regression", "broken", "error"	context-dependent — see Default below

Multi-Agent Combinations

Some tasks require more than one agent. Spawn in parallel when there are no dependencies:

Scenario	Agents
Full-stack feature	`backend-developer` + `frontend-developer` + optionally `database-specialist`
Security audit	`security-specialist` + `software-architect`
New API endpoint with UI	`backend-developer` → `frontend-developer` (sequential — API first)
Database schema + migration	`database-specialist` → `backend-developer` (sequential — schema first)
Test-only task (both layers)	`backend-test-specialist` + `frontend-test-specialist` (parallel)
PR review	`code-reviewer` + `security-specialist` + optionally `database-specialist`

Conflict Resolution

When a changed file matches multiple patterns:

Most specific pattern wins. src/controllers/UsersController.ts matches both frontend (src/) and backend (controller) — the controllers/ pattern is more specific, so spawn backend-developer.
When genuinely ambiguous, spawn software-architect first to classify and produce a plan.
Never spawn an agent for a domain not touched by the task — even if the agent is "usually helpful".

Default Rule

When scope is unclear — no strong file-path or text match — spawn software-architect first:

software-architect analyzes the task and produces a plan
Plan identifies which domains are affected
Downstream agents are spawned based on the plan output

This prevents incorrect agents from producing work that conflicts with the actual architecture.

Classification Checklist

Run git diff --name-only HEAD~1 to get changed files
Match files against path-pattern table above
Match task description against text-trigger table above
Identify parallel vs. sequential spawn order
If ambiguous, spawn software-architect first

How can the creator link this skill?

Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.

<a href="https://skillzs.dev/skills/dev-toolbelt/dev-team-agents/spawn-classifier">View spawn-classifier on skillZs</a>