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 rules/kyungseo/ai-workflow-harness/codinggit clone --depth 1 https://github.com/kyungseo/ai-workflow-harnessWhat 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 | $0.00604 | $0.00604 |
| Opus 5 | $0.00302 | $0.00302 |
| Sonnet 5 | $0.00121 | $0.00121 |
| Haiku 4.5 | $0.00060 | $0.00060 |
Grade A, and why
coding 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 2d 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.
What it actually says
Core Coding Rules
Instruction priority:
CLAUDE.md— global principlesdocs/AGENT-WORKFLOW.md— project operating rules.cursor/rules/*— tool-specific rules
MUST:
- Read the relevant project context before editing.
- Use
docs/STATUS.mdas the dashboard pointer view for active work. - Follow the workflow state machine:
INIT -> PLAN -> APPROVAL -> EXECUTE -> VALIDATE -> CHECKPOINT -> END. - Request explicit user approval before editing
docs/STATUS.md; report the Approval Matrix state-change proposal first. - Use
docs/BOOTSTRAP.mdonly whendocs/STATUS.mdNext Actions point to scaffold bootstrap/onboarding. - Use
docs/backlog/PRODUCT.mdfor Product track candidate work (optional phasing:PRODUCT-P{n}.md). - Use
docs/backlog/HARNESS.mdfor harness, command/rule, and workflow hardening candidate work. - Include scope, files, verification, risk, and reversal cost before implementation.
- Before expanding approved scope to additional files, docs, or settings, report the added scope, reason, and verification plan, then wait for approval.
- Use Work files for large tasks:
docs/works/{category}/{ID}-{lowercase-topic}.md(spec: DR-013). - When creating a new
docs/troubleshooting/ordocs/retrospectives/file, apply the DR-027 frontmatter spec defined in each directory'sREADME.md. - Treat Work files as the task SSoT and
docs/STATUS.mdas the dashboard pointer view. - Use Quick Mode for small Product track L1 changes that do not need Work file tracking; close with final summary, validation, and commit history. Treat harness/workflow surface changes as L2.
- Do not reuse task IDs for different meanings.
- Use path-mirrored locations under
docs/archive/only for historical detail. - State assumptions before implementation when the task is ambiguous.
- Keep every change SURGICAL, MINIMAL, and REVERSIBLE.
- Follow the existing package structure, naming, formatting, and coding conventions.
- Remove unused imports, variables, functions, and test fixtures introduced by your own change.
NEVER:
- Add unrequested features.
- Refactor unrelated code.
- Expand approved scope silently.
- Implement before plan approval when the change is non-trivial or user-facing.
- Rewrite adjacent code only for style preference.
- Introduce a new framework, dependency, or architectural layer without explicit justification and approval.
- Load long historical documents by default when the current state file is enough.
STOP AND ASK when requirements are unclear, multiple interpretations are possible, or the safer path depends on user intent.
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.
- 2d ago First seen · 48 lines · 604 tokens per session scan A 19c2fd63f9f7
coding is a cursor rule published in the GitHub repository kyungseo/ai-workflow-harness (13 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 604 tokens to every session, about $0.0030 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 2026-08-30.
Other cursor rules, from other repositories
language-agnostic-patterns
Language-agnostic programming patterns: SOLID, design patterns, clean code, and architecture. Load when refactoring, designing abstractions, or reviewing structure — not for everyday syntax.
cursor-tools-mastery
Cursor 3.7 runtime guide: choose the right tool, canvases, Design Mode, /worktree, /best-of-n, Await, and parallel execution where safe.
fable5-coding-craft
Fable 5 coding craft: locate-before-write, root-cause method, simplicity taste, error-handling philosophy, test integrity, refactoring discipline, and counters to common LLM coding failure modes. Load when writing, refactoring, debugging, or reviewing non-trivial code in any language.
cursor-agent-orchestration
Cursor 3.7 orchestration guide: when to plan, when to delegate, nested subagents, multi-environment handoffs, /best-of-n, and Await for long-running branches.
fable5-reasoning
Fable 5 reasoning protocols: task interpretation, risk-first decomposition, approach selection, interleaved thinking, hypothesis ledgers, premortems, calibration, and the stuck-strategy ladder. Load for complex, ambiguous, or long-horizon tasks, for debugging strategy, or whenever progress stalls.
cursor-mcp-optimization
Cursor 3.7 MCP optimization: browser Design Mode, canvases, Figma, Cloudflare tools, MCP Apps structured content, and direct action patterns.