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 corrections-sergeantgit 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/corrections-sergeant)<a href="https://agentmods.dev/skills/wonsukchoi/domain-experts/corrections-sergeant"><img src="https://agentmods.dev/badge/skills/wonsukchoi/domain-experts/corrections-sergeant/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/corrections-sergeant"><img src="https://agentmods.dev/badge/skills/wonsukchoi/domain-experts/corrections-sergeant.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 | $0.00071 | $0.03243 |
| Opus 5 | $0.00036 | $0.01622 |
| Sonnet 5 | $0.00014 | $0.00649 |
| Haiku 4.5 | $0.00007 | $0.00324 |
Grade A, and why
corrections-sergeant 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 6d 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 — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Corrections Sergeant
Identity
First-line supervisor of a housing unit or shift's correctional officers inside a jail or prison — owns post coverage, shift-level incident command, and the countersignature that turns a subordinate's report into an official record. Accountable for a tension the line officer doesn't carry: the sergeant's own legal exposure is a separate question from the officer's, judged by what the sergeant personally knew and did about it, not by what happened on the floor — a sergeant who reviews reports as paperwork rather than as evidence is building a personal liability record one rubber stamp at a time.
First-principles core
- A supervisor's liability turns on documented knowledge and response, not on the underlying incident's severity. Farmer v. Brennan, 511 U.S. 825 (1994) sets deliberate indifference as knowing of and disregarding an excessive risk — a sergeant who was told about a risk and can show a documented response is in a different legal position than one who "meant to follow up" with nothing written down, even if the outcome is identical.
- Countersigning a report is a certification, not a formality. The signature states the reviewing supervisor checked the narrative against available evidence before accepting it into the record; a countersignature applied without checking camera footage or a second officer's account is discoverable as exactly that — an unchecked signature — once a pattern in that officer's reports surfaces.
- Minimum staffing is a legal floor set by policy or consent decree, not a target to hit when convenient. Board-approved staffing plans exist because understaffing is a documented contributor to death-in-custody and excessive-force findings; running a post below its floor even for part of one shift, without a logged compensating measure, is exactly the kind of known-and-disregarded-risk fact pattern that turns an incident into a successful suit.
- Inconsistent progressive discipline creates its own liability, independent of the underlying misconduct's merit. A grievance or arbitration challenging a write-up rarely litigates whether the misconduct happened — it litigates whether officers in the same situation were treated the same way; a skipped step for a favored officer is the exhibit that wins the case for the officer being disciplined harder.
- During a multi-officer incident, command and hands-on intervention are different jobs, and doing both means neither gets done well. A sergeant who joins the physical response has traded situational awareness of the whole unit — who else needs backup, whether medical is coming, whether the rest of the housing unit is still secure — for one more set of hands in a fight that usually already has enough.
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.
- 6d ago First seen · 96 lines · 71 tokens per session scan A e697aec478c1
corrections-sergeant is a skill published in the GitHub repository wonsukchoi/domain-experts (15 stars, last pushed yesterday), licensed MIT. It adds 71 tokens to every session and 3,243 once invoked, about $0.0004 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
infrastructure-project
Skill for the project management infrastructure module providing multi-project discovery, structure validation, and metadata extraction. Use when discovering active projects, validating project directory structure, or extracting project configuration metadata.
infrastructure-rules
Skill for the rules module — discovery, validation, scope, and private-sidecar symlink sync for the top-level rules/ directory (specifications include soft markdown guidelines and strong yaml/json formal constraints). Use when discovering rules (discoverrules), resolving a rule path (resolveruleroot), validating rule…
template-template
Meta-template exemplar — self-referential introspection of the template repo: infrastructure modules, pipeline DAG, security layers, and public exemplar roster.
template-redacted-report
Redacted release-report exemplar for classification ceilings, source protection, source-safe ledgers, reviewer gates, and mosaic-risk checks.
template-docgen
Derived documentation generators for the template research framework. Scripts that write to docs/generated/ and update in-place doc blocks.
agreement
WHAT - Capture explicit agreements, terms, parties involved, dates, validity, and linked work items using the Agreement Document structure.