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 agentmods add instructions/tellang/triflux/gemini-mdgit clone --depth 1 https://github.com/tellang/trifluxWrote 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/instructions/tellang/triflux/gemini-md)<a href="https://agentmods.dev/instructions/tellang/triflux/gemini-md"><img src="https://agentmods.dev/badge/instructions/tellang/triflux/gemini-md.svg" alt="Measured on agentmods" 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 | $0.01008 | $0.01008 |
| Opus 5 | $0.00504 | $0.00504 |
| Sonnet 5 | $0.00202 | $0.00202 |
| Haiku 4.5 | $0.00101 | $0.00101 |
Grade A, and why
triflux GEMINI.md 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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
triflux — Gemini 프로젝트 지시
정책 SSOT는
.claude/rules/, 결정 근거는docs/adr/README.md.
triflux 워크스페이스에서 Antigravity CLI의 역할과 운영 규칙. 1M context 활용·웹 검색·헤드리스 합의의 한 축으로 사용된다.
전체 프로젝트 컨텍스트는 CLAUDE.md를 참조한다 (이 파일은 Gemini 관점만 다룸).
행동 규칙 (파일 상단 우선 적용)
- 한국어 우선, 기술 용어 원어 유지
- 응답은 결정론적·구조적으로 — 마크다운 헤더 또는 XML 태그 일관 사용 (혼용 금지)
- 부정 지시도 효과적 (긍정과 동등한 adherence — A/B 검증)
- 커밋:
Type: 한국어 설명 (50자 이내). Co-Authored-By / AI trailer 금지 - 시크릿(API 키, 토큰, 세션) 하드코딩 금지
Gemini의 역할
- 빠른 웹 검색 (Google Search 통합) —
tfx-research --quick - 3-CLI 합의(consensus/debate/panel)에서 Gemini 시점 제공
- 1M context 활용 — 큰 코드베이스 분석 /
tfx-interview는 Gemini 단독 - swarm/multi headless 병렬 실행의 워커
headless-guard
agy --dangerously-skip-permissions --print=직접 호출은 차단됨- 반드시 tfx 스킬 경유: tfx-auto (--cli antigravity), tfx-multi, tfx-swarm 등
- 일반 호출:
tfx-route.sh --cli antigravity ...(legacy--cli gemini는 agy 미가용 시 fallback)
디렉터리 import 금지 (EISDIR)
@./path/처럼 디렉터리 import하면 Gemini ImportProcessor가 fs.readFile을 호출해 EISDIR로 크래시 (gemini-cli #6450, #4760)- 디렉터리 경로를 문서에 남겨야 한다면 백틱(`)으로 감싸 import 스캔에서 제외
핵심 운영 컨텍스트 (요약 — 상세는 CLAUDE.md)
- psmux/WT 규칙:
tfx-psmux-rules스킬 / WT 프리징 방지 (exit → sleep 2 → kill) - AccountBroker 싱글턴: 계정별 CircuitBreaker, busy 플래그, 이벤트 기반 HUD 연동
- 원격 실행: tfx-remote (SSH → Claude Code → 내부 tfx 라우팅)
- Headless 결과 회수: task-notification 완료 후 output 파일 읽기
- 완료 마커:
=== HEADLESS_COMPLETE succeeded=N failed=N total=N ===
- 완료 마커:
교차 검증
- Claude 작성 코드 → Gemini/Codex 리뷰 가능
- Gemini 작성 코드 → Claude 또는 Codex 리뷰
- 동일 모델 self-approve 금지
검증
- 변경 후 lint/typecheck/test 실행
- 새 차단 규칙은 항상 대안 API와 함께 추가 (차단만 추가하면 데드락)
Skill routing
When the user's request matches an available skill, invoke it via the Skill tool. When in doubt, invoke the skill.
Key routing rules:
- Product ideas/brainstorming → invoke /gstack:office-hours
- Strategy/scope → invoke /gstack:plan-ceo-review
- Architecture → invoke /gstack:plan-eng-review
- Design system/plan review → invoke /gstack:design-consultation or /gstack:plan-design-review
- Full review pipeline → invoke /gstack:autoplan
- Bugs/errors → invoke /gstack:investigate
- QA/testing site behavior → invoke /gstack:qa or /gstack:qa-only
- Code review/diff check → invoke /gstack:review
- Visual polish → invoke /gstack:design-review
- Ship/deploy/PR → invoke /gstack:ship or /gstack:land-and-deploy
- Save progress → invoke /gstack:context-save
- Resume context → invoke /gstack:context-restore
- Author a backlog-ready spec/issue → invoke /gstack:spec
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 · 72 lines · 1,008 tokens per session scan A e878d63672b4
triflux GEMINI.md is an instructions file published in the GitHub repository tellang/triflux (7 stars, last pushed 3d ago), licensed MIT. It adds 1,008 tokens to every session, about $0.0050 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 2026-08-31.
Other instructions, from other repositories
neo4j-skills AGENTS.md
AGENTS.md instructions for neo4j-contrib/neo4j-skills, covering writing skills for this repository, language style — write for agents, not humans, six design principles, skill.md spec (agentskills.io) and frontmatter.
apm python.instructions.md
Python development guidelines.
OpenMicro AGENTS.md
Instructions for stephenleo/OpenMicro, covering openmicro — project conventions and releases.
shift-log CLAUDE.md
Claude Code instructions for re-cinq/shift-log, covering openspec instructions, integration tests and beads task tracking.
Lossless-Codex-Orchestrator-LCO AGENTS.md
AGENTS.md instructions for 100yenadmin/Lossless-Codex-Orchestrator-LCO, covering lco agent instructions, repository agent quick start, shared owned-repo policy, lco release gates and public documentation placement.
ai-dev-methodology-codex AGENTS.md
AGENTS.md instructions for stepanenkoviktor0110-boop/ai-dev-methodology-codex, covering global preferences, ⛔ rule #0: auto-update before any pipeline step, instruction tone: strict by default, skill execution: no scripts and recovery after interruption.