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 skills/sw-in-beta/simon-skills/simon-devnpx skills add SW-in-beta/simon-skills --skill simon-devgit clone --depth 1 https://github.com/SW-in-beta/simon-skillsWrote 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/sw-in-beta/simon-skills/simon-dev)<a href="https://agentmods.dev/skills/sw-in-beta/simon-skills/simon-dev"><img src="https://agentmods.dev/badge/skills/sw-in-beta/simon-skills/simon-dev.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 | $0.00170 | $0.12339 |
| Opus 5 | $0.00085 | $0.06169 |
| Sonnet 5 | $0.00034 | $0.02468 |
| Haiku 4.5 | $0.00017 | $0.01234 |
Grade A, and why
simon-dev 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 5d 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 — 494 lines — stays where its author put it; the contents beside it link to each section on GitHub.
simon-dev
Deep workflow skill with 19-step quality pipeline.
Instructions
You are executing the simon-dev deep workflow. This is a 19-step quality pipeline that plans, implements, and verifies code with maximum rigor.
요청이 심하게 막연하여 해석 자체가 갈리는 경우(목표·대상·형태가 모두 불특정), Phase A에 진입하지 말고 simon-resolve-unknowns 스킬로 발산 단계를 먼저 제안한다 (파이프라인: resolve-unknowns → plan → dev).
State-Driven Execution
매 턴(응답) 시작 시 반드시 실행하는 루틴 — compaction, 세션 재개, 작업 전환 후 복귀 등 어떤 상황에서도 적용된다. 이 루틴이 없으면 compaction 후 현재 위치를 잃고 이전 Step을 반복하거나 건너뛸 수 있다:
-
{SESSION_DIR}/memory/workflow-state.json읽기 -
current_step에 해당하는 Phase의 reference 목록 확인 (Reference Loading Policy 테이블) -
references_loaded필드에 해당 reference가 없거나, 있더라도loaded_at이 현재session_id이전이면 → 재로딩. 그 외에는 → 스킵. Compaction 감지는 결정론적 JSON 비교로 수행한다 — 아래 보조 조건 중 하나라도 해당하면 재로딩 트리거:- 조건 1: 세션 재개 (session_id 불일치 또는 없음)
- 조건 2:
/compact명령 직후 - 조건 3:
loaded_at이 현재session_id(ISO-8601) 이전인 Tier 1 항목이 존재 (compaction 소실 의심) - 조건 4: 매 턴 시작 시 Tier 1 파일 하나라도
references_loaded에 없음 (방어적 재로딩)
Compaction 감지 시 Tier별 선택적 초기화: Tier 1은 강제 재로딩, Tier 2는 현재 Phase에 해당하는 것만 재로딩, Tier 3는 초기화하지 않음 (on-demand 트리거로 자연스럽게 재로딩). "이전에 읽었으니 알고 있다"가 LLM 기억이 아닌 JSON 기록에 기반해야 핵심 규칙 소실을 방지한다.
원칙: "기억 판단 비용 > 1개 파일 재로딩 비용" — 의심스러우면 재로딩한다. 3-B. 이전 Step 산출물 로딩 (세션 독립성 핵심):
step_outputs[prev_step]에 기록된 파일들을 로딩한다.
- 새 세션 감지 시 (workflow-state.json의
session_id가 현재 Startup에서 설정한 값과 다르거나 없으면): 반드시 로딩 - 동일 세션 내: 이미 컨텍스트에 있으면 스킵
- Why: 새 세션에서 Step N을 시작할 때 Step N-1의 산출물 파일이 컨텍스트에 없으면 잘못된 판단이나 중복 작업이 발생한다.
step_outputs는 Step 완료 시 기록되므로, 이 파일들을 로딩하면 이전 Step의 결과를 즉시 복원할 수 있다.
- 해당 Step 실행
- Step 완료 즉시 workflow-state.json 갱신 (
references_loaded+step_outputs포함)
workflow-state.json이 없으면 Startup부터 시작한다.
갱신 규칙: Step 시작 시 current_step, 완료 시 completed_steps+next_step, Phase 전환 시 current_phase+phase_timestamps 갱신. 중단/에러: blocked: true. Step skip 시 skipped_steps 기록. Phase A 완료 시 Done-When Checks → done_when_checks 배열 추출 (verified: false 초기값). Step 5d/6/17 검증 통과 시 verified: true 갱신. Step 17에서 verified: false 잔존 시 FAIL — JSON boolean은 명시적 갱신이 필요하므로 LLM의 임의 체크 방지.
What ships with it
25 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.
- README.md 10 KB
- references/agent-capability-matrix.md 9.4 KB
- references/agent-teams.md 11 KB
- references/context-separation.md 9.3 KB
- references/cross-cutting-protocols.md 28 KB
- references/docs-first-protocol.md 3.0 KB
- references/error-resilience.md 12 KB
- references/expert-output-schema.md 11 KB
- references/forbidden-rules.md 4.9 KB
- references/gate-definitions.md 9.7 KB
- references/generation-style-guide.md 17 KB
- references/gotchas.md 3.8 KB
- references/handoff-manifest-brief.md 1.4 KB
- references/integration-and-review.md 20 KB
- references/openspec-capture.md 2.8 KB
- references/parallel-unit-orchestration.md 17 KB
- references/phase-a-planning.md 34 KB
- references/phase-a-review.md 26 KB
- references/phase-b-implementation.md 42 KB
- references/phase-b-verification.md 39 KB
- references/review-rubric.md 11 KB
- references/startup-bootstrap.md 1.1 KB
- scripts/create-branch.sh 1.2 KB runs code
- scripts/create-unit-worktree.sh 1.7 KB runs code
- scripts/merge-units.sh 2.1 KB 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.
- 5d ago First seen · 494 lines · 170 tokens per session scan A 5d95a0cb3c17
simon-dev is a skill published in the GitHub repository SW-in-beta/simon-skills (2 stars, last pushed 1mo ago), licensed MIT. It adds 170 tokens to every session and 12,339 once invoked, about $0.0009 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 skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…