Stats
Actions
Tags
From vibeflow
Validates proposed changes against architectural rules — layering, allowed dependencies, forbidden imports, naming conventions. Use before merging or when reviewing a refactor that touches cross-module boundaries. Blocks work that violates ADR-recorded constraints.
How this skill is triggered — by the user, by Claude, or both
Slash command
/vibeflow:arch-guardrailsThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Statically checks that a change respects the project's recorded architectural
Statically checks that a change respects the project's recorded architectural
rules. The rules live in .vibeflow/artifacts/arch-rules.yaml (or a path set
via vibeflow.config.json → archRulesPath). Absent rules mean "no
constraints"; the skill reports success in that case rather than inventing
defaults.
.vibeflow/artifacts/arch-rules.yaml — declared rules.vibeflow/traces/changed-files.log)
so the scan is incremental. Without one, the skill scans the whole tree.ui/ must not import from infra/).tests/).*.test.ts lives next to *.ts).Every finding follows the standard explainability shape:
{
"finding": "...rule violation summary...",
"why": "...which rule + which file/line...",
"impact": "blocks merge | soft warning | informational",
"confidence": 0.0
}
Write the findings list to .vibeflow/reports/arch-guardrails.md. Exit with
a non-zero code only when at least one finding has impact: blocks merge.
/vibeflow:advance from DESIGN → ARCHITECTURE (check preconditions)commit-guard.sh in Sprint 2)/vibeflow:arch-guardrailsnpx claudepluginhub mytechsonamy/vibeflow --plugin vibeflowProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.