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 Kminer2053/public-sector-research-skill --skill public-sector-researchgit clone --depth 1 https://github.com/Kminer2053/public-sector-research-skillWrote 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/kminer2053/public-sector-research-skill/public-sector-research)<a href="https://agentmods.dev/skills/kminer2053/public-sector-research-skill/public-sector-research"><img src="https://agentmods.dev/badge/skills/kminer2053/public-sector-research-skill/public-sector-research/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/kminer2053/public-sector-research-skill/public-sector-research"><img src="https://agentmods.dev/badge/skills/kminer2053/public-sector-research-skill/public-sector-research.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 | $0.00086 | $0.01304 |
| Opus 5.5 | $0.00034 | $0.00522 |
| Sonnet 5.5 | $0.00017 | $0.00261 |
| Haiku 4.5 | $0.00009 | $0.00130 |
Grade A, and why
public-sector-research 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 2mo 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 — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
공공복리 · BOKRI
공공복리(복리)는 공공업무 질문을 바로 요약하지 않고, 공식 원문을 확보한 뒤 근거 구간과 한계를 검토 가능한 형태로 남긴다. 이 Skill은 특정 AI 서비스의 검색 기능에 종속되지 않는다. 호스트가 제공하는 웹 검색·브라우저·커넥터로 공식 출처를 찾고, bundled Python CLI로 계획·수집·저장·점수화·보고서 생성을 수행한다.
실행 진입점
Skill 폴더의 scripts/psr.py를 사용한다.
python3 <skill-directory>/scripts/psr.py --version
설치형 CLI가 있으면 같은 계약의 psr 명령을 사용해도 된다.
기본 워크플로
- 사용자의 업무 목적, 기준일, 관할, 산출물을 확인한다.
- 프로젝트에
.psr/가 없으면 로컬 저장소를 초기화한다. research plan을 실행하고 생성된plan.json의 활성 track을 검토한다. 질문과 무관한 track은--exclude-track, 반드시 필요한 track은--include-track으로 조정한다.- 각 track의 query와 preferred domain을 사용해 공식 원문을 우선 탐색한다.
- 선택한 출처를 JSONL manifest에 기록한다. 출처 등급을 추측하지 말고 확인 가능한 범위에서 지정한다.
research run으로 출처를 병렬 수집·파싱하고 보수적인 초안 보고서를 생성한다.result.json의 citation을 검토한 뒤 reporting.md에 따라brief.json을 작성한다.report build --brief-file <path> --format all로 Markdown과 HTML을 함께 생성한다.PARTIAL, gap, failure, 발행일 미확인,UNVERIFIED_WEB항목을 먼저 검토한다.- 사용자에게 핵심 내용과 함께 HTML·Markdown 경로, 핵심 citation ID, 공식 원문 미확보 범위를 알린다.
빠른 시작
python3 <skill-directory>/scripts/psr.py project init . --name "AI 구매원칙"
python3 <skill-directory>/scripts/psr.py --project . research plan \
"공공기관 생성형 AI 구매 시 데이터 권리와 업체 종속 방지 원칙을 조사하라"
계획 출력의 run_id와 query를 사용해 출처를 찾은 뒤 sources.jsonl을 만든다.
{"track_id":"law-regulation","url":"https://official.example/law","title":"공식 법령","publisher":"공식기관","source_tier":"OFFICIAL_PRIMARY","published_at":"2026-01-01"}
{"track_id":"privacy","url":"https://official.example/privacy.pdf","title":"개인정보 안내서","publisher":"공식기관","source_tier":"OFFICIAL_PRIMARY","published_at":"2025-08-07"}
python3 <skill-directory>/scripts/psr.py --project . research run <run-id> \
--sources-file sources.jsonl
재사용과 오프라인 작업
동일 출처의 최근 snapshot은 기본 7일 동안 재사용한다. 최신성 확인이 필요하면 --refresh를 사용한다.
python3 <skill-directory>/scripts/psr.py --project . memory search "학습 재사용"
python3 <skill-directory>/scripts/psr.py --project . evidence list --run-id <run-id>
python3 <skill-directory>/scripts/psr.py --project . report build <run-id> \
--brief-file brief.json --format all
What ships with it
20 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.
- agents/openai.yaml 359 B
- references/evidence-policy.md 2.2 KB
- references/portability.md 1.6 KB
- references/reporting.md 3.7 KB
- references/research-profiles.md 3.0 KB
- references/source-manifest.md 1.3 KB
- references/workflow.md 3.9 KB
- scripts/psr_core/__init__.py 88 B runs code
- scripts/psr_core/briefing.py 7.2 KB runs code
- scripts/psr_core/cli.py 9.7 KB runs code
- scripts/psr_core/collector.py 9.9 KB runs code
- scripts/psr_core/evidence.py 11 KB runs code
- scripts/psr_core/models.py 4.8 KB runs code
- scripts/psr_core/parsers.py 13 KB runs code
- scripts/psr_core/planner.py 6.4 KB runs code
- scripts/psr_core/profiles.py 15 KB runs code
- scripts/psr_core/reporting.py 31 KB runs code
- scripts/psr_core/storage.py 27 KB runs code
- scripts/psr_core/workflow.py 14 KB runs code
- scripts/psr.py 396 B runs code
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.
- 2mo ago First seen · 88 lines · 86 tokens per session scan A fbf022a450ac
public-sector-research is a skill published in the GitHub repository Kminer2053/public-sector-research-skill (4 stars, last pushed 3mo ago), licensed MIT. It adds 86 tokens to every session and 1,304 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-08-31.
public-sector-research compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| public-sector-researchKminer2053/public-sector-research-skill | Skill | 4 | 3mo ago | MIT | A | 86 | Codex |
| tw-formal-writingImbad0202/tw-formal-writing | Skill | 197 | 1mo ago | MIT | A | 538 | Claude Code |
| policy-researchparkjui92/policy-research-kit | Skill | 10 | yesterday | MIT | A | 141 | Claude Code |
| federal-regulations-guidecyanheads/federal-regulations-mcp-server | Skill | 2 | 8d ago | Apache-2.0 | A | 51 | Claude Code |
| persona-lawyersodam-ai/SoDam-Persona | Skill | 17 | 2mo ago | Apache-2.0 | A | 74 | Claude Code |
| 052-Frank-G端合规方案撰写工具frank666199/frank-presales-skills | Skill | 23 | 4mo ago | MIT | A | 0 | Claude Code, Codex |
| 081-Frank-AI治理方案工具frank666199/frank-presales-skills | Skill | 23 | 4mo ago | MIT | A | 0 | Claude Code, Codex |