ln-646-project-structure-auditor — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited ln-646-project-structure-auditor (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Paths: File paths (references/,../ln-*) are relative to this skill directory.
Type: L3 Worker
L3 Worker that audits whether physical project structure reflects the intended architecture.
MOVE_MODULE, SPLIT_JUNK_DRAWER, or ALIGN_DOMAIN_STRUCTUREOut of Scope:
.gitignore, .dockerignore, temp files, build artifacts, platform remnants, and large binariesMANDATORY READ: Load references/audit_worker_core_contract.md. Tool policy: follow host AGENTS.md MCP preferences; load references/mcp_tool_preferences.md and references/mcp_integration_patterns.md only when host policy is absent or MCP behavior is unclear.
Receives contextStore with: tech_stack, architecture, codebase_root, output_dir, domain_mode, scan_path.
Use hex-graph first when architecture summaries materially improve structure findings. Use hex-line first for local code, config, and manifest reads when available. If MCP is unavailable, unsupported, or not indexed, continue with built-in Read/Grep/Glob/Bash and state the fallback in the report.
Detection policy: use two-layer detection (candidate scan, then context verification); load references/two_layer_detection.md only when the verification method is ambiguous.
1) Parse context -- determine scan_root from domain-aware scope or codebase root 2) Detect tech stack and intended structure
references/stack_detection.md3) Run structure checks (Layer 1)
4) Verify context (Layer 2)
5) Collect findings with severity, location, action, effort, and recommendation 6) Calculate score using references/audit_scoring.md 7) Write report to {output_dir}/ln-646--{identifier}.md 8) Return summary per references/audit_summary_contract.md
What: Source files are placed outside framework or stack conventions
Detection: Load references/structure_rules.md and apply only rules for the detected stack.
Severity: HIGH when placement breaks tooling/routing, MEDIUM for maintainability drift
Action: MOVE_MODULE
What: Domain or layer directories do not reflect the intended architecture
Detection:
Severity: MEDIUM by default, HIGH when structure hides critical ownership
Action: ALIGN_DOMAIN_STRUCTURE
What: Generic directories accumulate unrelated modules
Detection: Check directories such as utils, helpers, common, shared, services, and lib for mixed unrelated responsibilities above project thresholds.
Severity: MEDIUM for broad mixed responsibilities, LOW for small localized drift
Action: SPLIT_JUNK_DRAWER
MANDATORY READ: Load references/audit_scoring.md.
MANDATORY READ: Load references/templates/audit_worker_report_template.md.
Write JSON summary per references/audit_summary_contract.md. In managed mode the caller passes both runId and summaryArtifactPath; in standalone mode the worker generates its own run-scoped artifact path per shared contract.
Write report to {output_dir}/ln-646--{identifier}.md with category: "Project Structure" and checks: framework_placement, domain_layer_layout, junk_drawers.
When summaryArtifactPath is absent, write the standalone runtime summary under .hex-skills/runtime-artifacts/runs/{run_id}/evaluation-worker/{worker}--{identifier}.json and optionally echo the same summary in structured output.
Apply the already-loaded references/audit_worker_core_contract.md.
MOVE_MODULE, SPLIT_JUNK_DRAWER, or ALIGN_DOMAIN_STRUCTURE.Apply the already-loaded references/audit_worker_core_contract.md.
references/audit_scoring.md{output_dir}/ln-646--{identifier}.md (atomic single Write call)Version: 1.0.0 Last Updated: 2026-03-15
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.