simon-grind

simon-grind is a skill for Claude Code, Codex from SW-in-beta/simon-skills. It costs 200 tokens per session (5,133 once invoked), scanned A, original, MIT.

An automated coding workflow for retrying failed work, diagnosing errors, changing approach, and saving checkpoints. It is intended for tasks where giving up quickly is not acceptable.

In plain words
What is it for?
Use it for high-risk features, complex codebases with repeated build or test failures, or work that should continue with little human intervention. It is also suited to retrying tasks that another workflow could not finish.
Why use it?
It reduces the need to watch and restart a difficult task after build or test failures. It also records failures and progress so repeated attempts can be evaluated safely.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: reads .claude/ paths; mentions subagents; names the AskUserQuestion tool.

Good fit Use it for high-risk features, complex codebases with repeated build or test failures, or work that should continue with little human intervention. It is also suited to retrying tasks that another workflow could not finish.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/sw-in-beta/simon-skills/simon-grind
Install

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.

Any agent
npx skills add SW-in-beta/simon-skills --skill simon-grind
Clone the repo
git clone --depth 1 https://github.com/SW-in-beta/simon-skills

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for simon-grind

README.md
[![agentmods](https://agentmods.dev/badge/skills/sw-in-beta/simon-skills/simon-grind/github.svg)](https://agentmods.dev/skills/sw-in-beta/simon-skills/simon-grind)
Your own site
<a href="https://agentmods.dev/skills/sw-in-beta/simon-skills/simon-grind"><img src="https://agentmods.dev/badge/skills/sw-in-beta/simon-skills/simon-grind/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.

agentmods 80×15 button for simon-grind

Your own site · 80×15
<a href="https://agentmods.dev/skills/sw-in-beta/simon-skills/simon-grind"><img src="https://agentmods.dev/badge/skills/sw-in-beta/simon-skills/simon-grind.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 200 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,133 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00200 $0.05133
Opus 5 $0.00100 $0.02567
Sonnet 5 $0.00040 $0.01027
Haiku 4.5 $0.00020 $0.00513

Measured 10d ago against content hash 33485a2fe1be, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

simon-grind 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 10d ago.

The scan reads SKILL.md. This mod also ships 4 executable files (scripts/budget-tracker.sh, scripts/checkpoint.sh, scripts/done-when-eval.sh, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

skills/simon-grind/SKILL.md · 281 lines

How it starts

The opening of the file, as written. The whole thing — 281 lines — stays where its author put it; the contents beside it link to each section on GitHub.

simon-grind

열일모드 — 끝날 때까지 끝난 게 아니다.

Base Workflow

이 스킬은 simon-dev 19-step pipeline을 기반으로 합니다. 기본 워크플로는 ~/.claude/skills/simon-dev/SKILL.md와 그 references/ 파일들을 따릅니다.

이 문서는 grind 전용 확장사항만 기술합니다:

  • 모든 재시도 한계 = 10 (포기하지 않는다)
  • 새 재시도 지점 at previously non-retrying steps
  • 자동 진단 시스템 — failure tracking + strategy pivots
  • 체크포인트/롤백 for safe experimentation
  • Confidence Scoring & Assumption Registry — structured user interviews

Shared Protocols: ~/.claude/skills/_shared/preamble.md 읽기 — Session Isolation, Error Resilience, Forbidden Rules, Agent Teams, Cognitive Independence 공통 프로토콜 포함.

Session Isolation Protocol (확장)

grind 전용 파일(failure-log.md, checkpoints.md, assumptions-registry.md, escalation-report.md)도 동일하게 {SESSION_DIR}/memory/에 저장한다.

Grind Scripts

scripts/ 디렉토리의 결정론적 스크립트로 grind 핵심 메커니즘을 강제한다. LLM 기억에 의존하면 compaction 후 상태를 잃을 수 있으므로, 스크립트 출력에 기반하여 판단한다.

  • progress-detect.sh <failure-log> — 최근 2건 재시도 비교 → PROGRESS/STALL/REGRESS JSON 출력. Stall detection의 판정 근거로 사용.
  • budget-tracker.sh <failure-log> [budget] — 잔여 예산 JSON 출력. 70%+ 소비 시 warning: true. 매 재시도 시작 전에 실행하여 예산 상태를 확인한다.
  • checkpoint.sh <step> <attempt> <checkpoints-md>git tag checkpoint-step{N}-attempt{M} + checkpoints.md 원자적 갱신. 전략 전환 전에 반드시 실행한다.
  • done-when-eval.sh <plan-summary> <verify-commands> [unit] — Mechanical Done-When Checks 자동 평가 → exit_loop JSON 출력. 매 attempt 완료 시 실행하여 exit_loop=true면 해당 Step 루프를 조기 탈출한다. 성공 기준이 이미 충족됐는데 모르고 재시도하는 budget 낭비를 방지한다 (Behavioral Checks는 LLM 평가 유지).

Grind Config Overrides

loop_limits:
  # ── Existing (ALL → 10) ──────────────────────
  critic_planner: 10            # 3 → 10  (Step 2)
  step4b_critical: 10           # 2 → 10  (Step 4-B)
  step7b_recheck: 10            # 1 → 10  (Step 7-B)
  step7_8: 10                   # 2 → 10  (Step 7-8)
  step16: 10                    # 3 → 10  (Step 16)
  step6_executor: 10            # 3 → 10  (Step 6)

  # ── New Retry Points (ALL = 10) ──────────────
  step5_build_retry: 10         # build/test failure retry
  step5_strategy_pivot: 10      # full strategy pivot
  step9_11_refactor: 10         # refactoring loop
  step12_major: 10              # review loop
  step17_readiness: 10          # production readiness
  step19_review: 10             # simon-code-review 피드백 루프 내 재시도
  ci_watch_fix: 10              # CI failure auto-fix cycles
  expert_consensus_extra: 10    # extra discussion rounds
  integration_conflict: 10      # conflict resolution
  test_env_setup: 10            # test environment setup
  step13_cleanup: 10            # dead code cleanup
  step14_quality: 10            # quality improvement
  step15_flow: 10               # flow verification

  # ── Interview Protocol ───────────────────────
  interview_max_rounds: 5
  confidence_threshold: 0.85
  step1a_verification: true
  step2_4_escalation: true
  step4b_interactive: true
  max_questions_per_round: 5
  # 질문 정렬: 계획 구조를 바꿀 가능성이 큰 질문 우선. 상위 답변에 종속되는 질문은 같은 라운드에 묶지 않는다 (interview-protocol.md 번들 정렬 규칙과 동일).

  # ── Auto-Diagnosis ──────────────────────────
  diagnosis_threshold: 5        # Failures before root cause analysis
  pivot_threshold: 10           # Strategy pivots before human escalation

  # ── Progress Detection ─────────────────────
  stall_threshold: 2            # 진전 없는 연속 재시도 N회 시 전략 즉시 전환
  # 조정 가이드: 단순 버그 수정은 2(기본값) 유지.
  # 아키텍처 수준 변경이나 복잡한 통합 작업은 3-4로 상향 —
  # 이런 작업은 3~4회차에 누적 이해로 해결되는 경우가 있다.

  # ── Total Retry Budget ─────────────────────
  total_retry_budget: 50        # 워크플로 전체 재시도 예산
  # 조정 가이드: STANDARD는 50(기본값), LARGE는 80.
  # 외부 API 연동이 많은 작업은 +20 가산 — 환경 변수에 의한 실패가 빈번.
  budget_warning_pct: 70        # 예산 N% 소비 시 사용자 경고

  # ── Checkpoint Policy ─────────────────────
  checkpoint_policy:
    tier_boundary: true    # Attempt 3→4, 6→7 전환 시 자동 checkpoint
    on_progress: true      # Progress Detection "진전 있음" 시 best 태그
    strategy_pivot: true   # 기존과 동일
    rollback_target: best  # pivot 시 initial 대신 best로 롤백 (없으면 initial)

  # ── Per-Attempt Efficiency ───────────────
  per_attempt_warning_turns: 30   # 개별 attempt가 30턴 이상 소비 시 경고
  total_turns_warning: 200        # 전체 누적 턴이 200 이상 시 사용자에게 경고

Read the full file on GitHub · 281 lines

Files

What ships with it

8 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.

Changes

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.

  1. 10d ago First seen · 281 lines · 200 tokens per session scan A 33485a2fe1be

Subscribe to this mod's changes

simon-grind is a skill published in the GitHub repository SW-in-beta/simon-skills (2 stars, last pushed 1mo ago), licensed MIT. It adds 200 tokens to every session and 5,133 once invoked, about $0.0010 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.

Related

Other skills, from other repositories

debug-optimize-lcp

Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…

ChromeDevTools/chrome-devtools-mcp · 99 tokens

systematic-debugging

Use when debugging a failing test, build error, or runtime issue that isn't immediately obvious. Guides a 4-phase root cause analysis instead of random fix attempts.

open-metadata/OpenMetadata · 37 tokens

diagnose

Trace from a reproduced symptom to the source code that causes it. Pin the specific file and approximate line, rate confidence in the cause and clarity of the fix independently, and always propose a concrete fix.

emdash-cms/emdash · 43 tokens

repro-admin

Reproduce an EmDash admin UI bug. Attach a container, start the demo dev server, drive the admin with agent-browser using the dev-bypass session, and capture the reproduction as screenshots plus a replayable transcript.

emdash-cms/emdash · 48 tokens

log-error-digest

Analyze log files to troubleshoot errors, identify peak error periods, and produce error clustering, frequency statistics, and time distribution reports. Supports JSON, syslog, and Nginx formats with automatic detection. Use when a user uploads a .log file and asks to analyze errors, find patterns, debug issues, or…

zebbern/claude-code-guide · 71 tokens

byted-util-volcengine-detect-retry

An orchestration workflow for Volcengine Cloud Detect, a service that checks websites or network endpoints from test locations.

bytedance/agentkit-samples · 101 tokens