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 komluk/scaffolding --skill spec-designgit clone --depth 1 https://github.com/komluk/scaffoldingWrote 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/komluk/scaffolding/spec-design)<a href="https://agentmods.dev/skills/komluk/scaffolding/spec-design"><img src="https://agentmods.dev/badge/skills/komluk/scaffolding/spec-design.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 | $0.00051 | $0.00925 |
| Opus 5 | $0.00026 | $0.00463 |
| Sonnet 5 | $0.00010 | $0.00185 |
| Haiku 4.5 | $0.00005 | $0.00093 |
Grade A, and why
spec-design 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 7d 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 — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.
OpenSpec Design & Tasks Writing
Guide for creating design.md (WHAT + HOW) and tasks.md (implementation checklist).
Prerequisite
Read {specs_path}/proposal.md first. Reference its capabilities and impact sections.
Path Enforcement: The specs_path MUST be .scaffolding/conversations/{UUID}/specs/ where {UUID} is a valid UUID (format: xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx). NEVER use descriptive folder names.
design.md
Output Path
Write to: {specs_path}/design.md
Required Sections
| Section | Purpose | Format |
|---|---|---|
| Context | Background, current state, constraints | Prose |
| Goals / Non-Goals | Scope boundaries | Bullet lists |
| Specifications | Requirements + scenarios | GIVEN/WHEN/THEN |
| Decisions | Technical choices + rationale | Decision: Why X over Y |
| Risks / Trade-offs | Known limitations | [Risk] -> Mitigation |
| Migration Plan | Deploy + rollback steps | Checklist (if applicable) |
| Open Questions | Unresolved items | Numbered list |
Specification Format
### Requirement: user-session-timeout
The system SHALL terminate idle sessions after 30 minutes.
#### Scenario: session expires after inactivity
- **GIVEN** a user has an active session
- **WHEN** no activity occurs for 30 minutes
- **THEN** the session is invalidated and user is redirected to login
Rules:
- Use SHALL/MUST for normative requirements (avoid should/may)
- Every requirement MUST have at least one scenario
- Each scenario is a potential test case
- Map capabilities from proposal.md to requirement groups
Decision Format
## Decisions
### Use Redis for session storage
**Alternatives**: PostgreSQL, in-memory
**Rationale**: Sub-millisecond lookups, built-in TTL, already in stack
Scaffolding Constraints
Include in all designs:
- Max 200 lines per Edit operation
- Files must stay under 500 lines
- Backend validation:
pytest; Frontend:npm run validate - Map steps to specific workflow chain agents
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.
- 7d ago First seen · 118 lines · 51 tokens per session scan A ec1df5ce4940
spec-design is a skill published in the GitHub repository komluk/scaffolding (15 stars, last pushed 1mo ago), licensed MIT. It adds 51 tokens to every session and 925 once invoked, about $0.0003 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 skills, from other repositories
turnstile-loop-start
Starts a goal-directed loop run -- asks for mode, initialises state.json, then dispatches loop-runner repeatedly until the success condition is met, max-iterations is reached, or an unrecoverable error occurs.
turnstile-loop-status
Reads turnstile/loops/ /state.json and renders the current phase, status, and full iteration history in the terminal.
turnstile-conversate
Entry point for natural-language turnstile requests -- classifies the message against board state, then answers directly or invokes the one matching skill (brainstorm, quick, refine, spec, sprint-plan, work, review, breakdown, drop, systematic-debugger, code-reviewer).
turnstile-work
Implements one sprint ticket via TDD and the turnstile-coder agent, checking the brain first. Dispatched by /turnstile:work or conversate routing only.
turnstile-brain-init
Opt-in bulk bootstrap of turnstile/code/ -- one linked note per source file. The default pipeline is lazy (code notes only for ticket-touched files, written when the ticket passes review); this is the eager exception. Dispatched by /turnstile:brain-init only.
turnstile-brain
Vault layout, the shared note format, and the check-the-brain-first mandate. Auto-loads whenever any turnstile skill starts new work or /turnstile:review is about to commit.