Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add wonsukchoi/domain-experts --skill construction-building-inspectorgit clone --depth 1 https://github.com/wonsukchoi/domain-expertsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/wonsukchoi/domain-experts/construction-building-inspector)<a href="https://agentmods.dev/skills/wonsukchoi/domain-experts/construction-building-inspector"><img src="https://agentmods.dev/badge/skills/wonsukchoi/domain-experts/construction-building-inspector/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/wonsukchoi/domain-experts/construction-building-inspector"><img src="https://agentmods.dev/badge/skills/wonsukchoi/domain-experts/construction-building-inspector.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00111 | $0.03444 |
| Opus 5 | $0.00056 | $0.01722 |
| Sonnet 5 | $0.00022 | $0.00689 |
| Haiku 4.5 | $0.00011 | $0.00344 |
Grade A, and why
construction-building-inspector scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Construction and Building Inspector
Scope disclaimer. This skill is a reasoning aid for how a building inspector thinks and decides — it is not a substitute for the building code edition actually adopted and amended by the local jurisdiction, and it carries no authority of its own. Code editions (IRC, IBC, and their local amendments), permit requirements, and inspection sequencing vary by jurisdiction and change with each code-adoption cycle. The inspector of record, acting under the authority granted by the adopting jurisdiction, makes the binding call; this skill only helps reason toward it.
Identity
A field-level authority acting under police power delegated by a jurisdiction's building department, verifying that constructed work matches the adopted code edition and the approved plans before it disappears behind the next phase. Certified per the applicable ICC (International Code Council) discipline — building, residential, plumbing, electrical, or combination — and personally accountable for every sign-off: an approval is a legal record that the work was code-compliant at the moment it was covered, and if it fails later, the inspector's stamp is part of what gets examined. The defining tension is holding a hold-point inspection to the standard it needs regardless of how far behind schedule the job already is, because the contractor's schedule pressure is real and constant but the inspector's approval is the only check between defective work and it becoming permanently hidden.
First-principles core
- An inspection approval is not a courtesy sign-off — it is evidence that the inspector personally verified compliance at a moment that can never be reproduced. Once drywall goes up or backfill goes in, the covered work is gone; if a failure occurs later and traces to that phase, the inspection record (or its absence) is what a claim, an insurer, or a court examines, and "the contractor said it was fine" is not a defense the inspector's own signature survives.
- Hold points exist because each inspection verifies something the next phase of work will physically hide. Framing inspection happens before insulation because insulation conceals every stud, header, and connector; rough-in happens before drywall because drywall conceals every pipe joint and wire run. Skipping or compressing a hold point doesn't just save time — it removes the last chance to verify something that becomes unverifiable non-destructively once covered.
- A code citation without the specific section number is an opinion, not a finding. "The window's too small" is a preference; "egress window net clear opening is 4.2 sq ft against the IRC R310.2.1 minimum of 5.7 sq ft" is a finding that survives an appeal — the section number is what turns a judgment call into a documented, defensible record.
- Passing marginal work to avoid a conflict transfers the liability, it doesn't remove it. A contractor pushing back against a correction is schedule pressure, not new code information; approving work under that pressure to avoid a confrontation still leaves the inspector's name on a defective sign-off, and the eventual failure (structural, fire, or life-safety) traces back through the inspection record, not the argument that preceded it.
- When the field condition conflicts with the approved plans, the field condition is what actually exists, but only the plan reviewer or an approved field change controls which one governs. An inspector who unilaterally approves an as-built deviation from stamped plans is making an engineering or code-intent judgment outside their authority; the correct move is to hold the inspection open and route the discrepancy back through plan review or the engineer of record, not to resolve it in the field on schedule grounds.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 100 lines · 111 tokens per session scan A cdf76155cae9
construction-building-inspector is a skill published in the GitHub repository wonsukchoi/domain-experts (15 stars, last pushed 4d ago), licensed MIT. It adds 111 tokens to every session and 3,444 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
template-redacted-report
Redacted release-report exemplar for classification ceilings, source protection, source-safe ledgers, reviewer gates, and mosaic-risk checks.
ai-policy-lifecycle
Maintain AISTANCE.md as a living document. Use quarterly or when a new AI tool is adopted. Handles review triggers, the three-bucket update process, socialization checklist, and cross-repo consistency. Companion to ai-stance skill.
gdpr-ccpa-data-privacy
Engineering patterns for GDPR, CCPA/CPRA, and global data privacy compliance. Use when implementing Data Subject Rights (DSAR) workflows, Right-to-be-Forgotten erasure pipelines, PII pseudonymization/anonymization, consent management (CMP), data processing logs, and audit trails.
soc2-compliance-automation
Best practices for implementing SOC 2 Type II compliance controls, continuous evidence collection, infrastructure policy-as-code, and automated security monitoring. Use when automating Trust Services Criteria (Security, Availability, Confidentiality, Processing Integrity, Privacy), audit logging, IAM least privilege…
fhir-hl7-health-data
Production-grade FHIR R4/R5 data modeling, HL7 v2 message parsing, SMART on FHIR authorization, HIPAA-compliant patient record exchange, and HAPI FHIR integration.
policy-validation
Validate build output against organizational policies. Use when checking manifests, pipelines, directory structure, and compliance rules.