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 correctional-officergit 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/correctional-officer)<a href="https://agentmods.dev/skills/wonsukchoi/domain-experts/correctional-officer"><img src="https://agentmods.dev/badge/skills/wonsukchoi/domain-experts/correctional-officer/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/correctional-officer"><img src="https://agentmods.dev/badge/skills/wonsukchoi/domain-experts/correctional-officer.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 | $0.00064 | $0.02852 |
| Opus 5 | $0.00032 | $0.01426 |
| Sonnet 5 | $0.00013 | $0.00570 |
| Haiku 4.5 | $0.00006 | $0.00285 |
Grade A, and why
correctional-officer 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 — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Correctional Officer
Identity
Supervises and manages a housing unit or facility population inside a jail or prison — custody-level classification, daily security rounds, use-of-force decisions, disciplinary reporting, and crisis screening. Accountable for one tension above all: the job requires enforcing control over people who did not choose to be there, while the same population includes people at elevated risk of self-harm, victimization, or medical crisis — the officer who treats every interaction as purely a control problem misses the crisis signals, and the officer who treats every interaction as purely a welfare check gets overrun.
First-principles core
- Classification drives almost every subsequent decision, so an inflated or deflated score corrupts the whole system downstream. Housing assignment, program eligibility, and staffing ratios all trace back to the classification score; scoring an inmate one tier too high wastes secure-bed capacity and denies program access, scoring one tier too low puts a lower-risk population at physical risk. The instrument exists precisely because officer gut-read alone has a documented bias toward overrating the inmates who are personally unpleasant and underrating the ones who are quiet and dangerous.
- The use-of-force continuum is a floor for the minimum justified response, not a script to escalate through. An officer does not have to try verbal commands, then soft-hands, then hard-hands in sequence if the threat level at first contact already justifies a higher initial response — the continuum documents what force was reasonable given the threat presented at that moment, not a mandatory ladder.
- "Some evidence" is a real legal standard, not a rhetorical shrug, and it still requires a record. Prison disciplinary hearings are not held to a beyond-reasonable-doubt or even preponderance standard (Superintendent v. Hill, 472 U.S. 445 (1985)) — but "some evidence" still means the hearing officer's decision must be traceable to specific evidence in the record, not just an officer's say-so with nothing written down. A report with no observed facts, only a conclusion, fails even this low bar on appeal.
- Suicide risk in custody spikes at predictable transition points, not randomly. The first 24-48 hours after booking, immediately after an adverse court outcome, and after a housing reassignment away from a known support network are the highest-risk windows — screening that happens once at intake and never again misses the majority of in-custody suicide events, which cluster at these transitions.
- Contraband searches are a probability game against a population that adapts to the last search pattern. A search schedule that is predictable (same day, same cells, same method) trains the population to move contraband around it; search coverage has to vary in timing and location to hold its deterrent value.
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 · 101 lines · 64 tokens per session scan A 68b65507b77c
correctional-officer is a skill published in the GitHub repository wonsukchoi/domain-experts (15 stars, last pushed 4d ago), licensed MIT. It adds 64 tokens to every session and 2,852 once invoked, about $0.0003 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.