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/oriolshhh/runware-image-mcp/implement-spec-frontendgit clone --depth 1 https://github.com/Oriolshhh/runware-image-mcpWrote 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/commands/oriolshhh/runware-image-mcp/implement-spec-frontend)<a href="https://agentmods.dev/commands/oriolshhh/runware-image-mcp/implement-spec-frontend"><img src="https://agentmods.dev/badge/commands/oriolshhh/runware-image-mcp/implement-spec-frontend.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 | $0.00034 | $0.01495 |
| Opus 5 | $0.00017 | $0.00747 |
| Sonnet 5 | $0.00007 | $0.00299 |
| Haiku 4.5 | $0.00003 | $0.00150 |
Grade A, and why
implement-spec-frontend 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 — 125 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/implement-spec-frontend — Implement an approved frontend redesign
Purpose
Implement an approved /spec-frontend artifact faithfully, then prove the
redesign in the rendered product with independent specialist review.
Invocation
/implement-spec-frontend <spec name or path>
Accepted input
An approved .agent/specs/*.md file with work_type: frontend-redesign,
baseline conditions, an implementation plan, and browser acceptance scenarios.
Prerequisites
- The spec exists, is readable, and has
status: approved. - The target route, preview, stable fixture/data, and required source paths are available.
- Material brand, product, information-architecture, dependency, and shared design-system decisions are already approved.
Procedure
- Read the entire spec and confirm
generated_by: spec,work_type: frontend-redesign,status: approved, repository revision, source paths, preservation constraints, requirements, tasks, tests, browser scenarios, and approval-sensitive decisions. Stop if the spec is unapproved. - Apply
context-discovery, then verify change-sensitive claims against source. If the target, design system, route behavior, fixtures, or relevant source has materially drifted, return to/spec-frontendfor an amendment; do not reinterpret a stale direction while coding. - Reproduce the spec's baseline before editing with
browser-visual-validation. Use identical route, data, state, viewport, theme, and interaction conditions. If comparison conditions changed, repair the fixture or stop—the final cannot be compared against a different baseline. - Load and apply the curated canonical skills required by the approved scope:
frontend-guidelines,code-quality-guidelines,frontend-design-direction,visual-craft,design-system-integration,ui-state-design,responsive-ui-engineering,accessible-ui-engineering,motion-design,test-design, andbrowser-visual-validation. Loadarchitecture-guidelinesonly for approved shared frontend architecture, routing, data-flow, or design-system ownership changes. - Inspect active packs. If more than one primary aesthetic pack is enabled,
stop and ask the user to choose exactly one; do not rely on priority order to
reconcile contradictory visual directions. Otherwise apply at most one
explicitly enabled primary aesthetic pack
(
frontend-design,taste-skill, orui-ux-pro-max) plus optionalemil-design-engineering. Do not silently enable packs, blend conflicts, fetch remote prompts, execute upstream installers/scripts, or add a dependency because a third-party skill prefers it. - Hand the approved brief, context capsule, reuse map, preservation list,
tasks, and acceptance matrix to
frontend-implementer. Use one coherent production-code writer by default; do not assign overlapping UI files to parallel agents. Parallelize only truly independent read-only research, isolated tests, or non-overlapping assets explicitly marked safe by the spec. - Implement in dependency order: semantics and content, state/behavior, structure and responsive transitions, visual system, then justified interaction/motion. Reuse local primitives, keep scope coherent, and add the specified unit/component/E2E/visual/accessibility tests.
- Run independent final checks, in parallel when native isolated subagents are
available and otherwise as honestly labelled separated passes:
visual-qa-tester: like-for-like baseline/final browser comparison, relevant viewports/states/themes/zoom/long content, console, and network;accessibility-reviewer: keyboard, focus, semantics, names/states, contrast, reflow, reduced motion, forms, and recovery;visual-craft-specialist: approved direction, hierarchy, typography, color, spacing, density, alignment, shape, icon/image, and anti-slop fidelity;ux-reviewer: named task outcome, comprehension, friction, copy, and recovery;motion-design-reviewerwhen interaction or motion changed: purpose, responsiveness, interruption, reduced motion, and performance.
- Consolidate only reproducible, in-scope blockers. Have
frontend-implementerperform at most one correction pass, then re-run the affected checks. Do not create a hidden unbounded loop. If blockers remain, stop with evidence and use the configurableimprove-ui-uxloop only when the user explicitly wants additional bounded iterations. - Run configured gates. Write or update
.agent/evidence/ui/<spec-name>/report.mdwith baseline/final conditions, evidence paths, changed files, requirement coverage, specialist execution mode and verdicts, states/viewports, accessibility, motion, console/network, tests/gates, deviations, remaining blockers, and untested coverage.
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 · 125 lines · 34 tokens per session scan A 1d315f21071e
implement-spec-frontend is a command published in the GitHub repository Oriolshhh/runware-image-mcp (0 stars, last pushed 1mo ago), licensed MIT. It adds 34 tokens to every session and 1,495 once invoked, about $0.0002 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
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
constitution
Create or update the project constitution from interactive or provided principle inputs.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.