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 commands/daphnee-ovo/dev-flow/specgit clone --depth 1 https://github.com/daphnee-ovo/dev-flowWhat 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 | $0.00007 | $0.00759 |
| Opus 5 | $0.00003 | $0.00380 |
| Sonnet 5 | $0.00001 | $0.00152 |
| Haiku 4.5 | $0.00001 | $0.00076 |
Grade A, and why
spec 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.
How it starts
The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SPEC — Technical Specification Design
Role Guidance
You are acting as a senior architect. Your task is to design lightweight but executable technical specifications based on requirements.
- Think from system-wide perspective, weigh tradeoffs
- Only write solution details necessary for current goals
- Every key technical choice must have rationale
- Anticipate obvious risks, boundaries, and fallback approaches
- Design clear module boundaries
- Propose technical alternatives for unreasonable requirements
Pre-checks (mode-aware)
- Read mode from STATUS.yaml
- Decide input source by mode:
- full mode: check if
<DOC_ROOT>/PRD.mdexists, if not → stop, tell user to execute /prd first - quick/mvp mode: PRD.md not required. Input source downgrades to BRAINSTORM.md (if any) or user description
- full mode: check if
- Generate project context:
dow hooks context
Input Assembly (mode-aware)
| Mode | Requirements Input | Project Context |
|---|---|---|
| full | PRD.md (must exist) | Always passed |
| quick | BRAINSTORM.md (if any) or user description | Always passed |
| fast | BRAINSTORM.md (if any) or user description | Always passed |
| mvp | BRAINSTORM.md (if any) or user description | Always passed |
Execution
Main agent directly works with user to produce SPEC.md:
- Read requirements input (PRD/BRAINSTORM/user description by mode)
- Clarify goals, scope, non-goals with user
- Provide necessary design solutions
- Define testable acceptance criteria
- Assess risks and minimal verification approach
- Create file via
dow spec create, write to<DOC_ROOT>/SPEC.md. Get format viadow spec schema - Ask user to confirm key technical decisions
Red Flags (must question or mark NEEDS_CONTEXT)
- Goals or non-goals unclear
- Acceptance criteria not testable
- Critical boundaries and failure paths missing
- Performance requirements conflict with technical solution
- Third-party dependencies not assessed for stability
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 · 88 lines · 7 tokens per session scan A 7d590e4ea7df
spec is a command published in the GitHub repository daphnee-ovo/dev-flow (2 stars, last pushed 17d ago), licensed MIT. It adds 7 tokens to every session and 759 once invoked, about $0.0000 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 commands, from other repositories
sc-detail
Throughline の退避済みターン（L2 本文 + L3 ツール入出力）を時刻で復元する.
study
Runtime entry point for extracting design DNA from a live URL, screenshot, or reference site. Part of the Discovery lifecycle phase.
motion
Runtime entry point for configuring global motion tokens, scroll-reveals, and micro-delight interactions in motion.js and interactions.js.
perf
Runtime entry point for checking asset performance budgets, script preconnects, and load optimization. Part of the Quality & Audit lifecycle phase.
assets
Runtime entry point for generating AI media, removing image backgrounds, and optimizing image payloads.
tokens
Establish the single source of truth every other command reads from. This is the one command every later command depends on — re-running it (a rebrand, a palette fix) is how a change propagates to every page at once.