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 ucsandman/marketing-studio --skill de-vibegit clone --depth 1 https://github.com/ucsandman/marketing-studioWrote 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/ucsandman/marketing-studio/de-vibe)<a href="https://agentmods.dev/skills/ucsandman/marketing-studio/de-vibe"><img src="https://agentmods.dev/badge/skills/ucsandman/marketing-studio/de-vibe/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.
<a href="https://agentmods.dev/skills/ucsandman/marketing-studio/de-vibe"><img src="https://agentmods.dev/badge/skills/ucsandman/marketing-studio/de-vibe.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 63 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00128 | $0.01898 |
| Opus 5 | $0.00064 | $0.00949 |
| Sonnet 5 | $0.00026 | $0.00380 |
| Haiku 4.5 | $0.00013 | $0.00190 |
Grade A, and why
de-vibe 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 11d 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 — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
De-Vibe
Vibe-coded software has a recognizable fingerprint: security holes nobody checked (wildcard CORS, committed .env, client-side-only auth), reliability debris (empty catches, console.log spam, tautological tests), the "AI slop" aesthetic (indigo gradients, Inter everywhere, untouched shadcn defaults, "Elevate your workflow" copy), and repo artifacts that announce how it was built. This skill removes that fingerprint and replaces the generic defaults with a small, deliberate identity — so the shipped product reads as handcrafted, and more importantly, doesn't ship with the real defects that make vibe-coded apps break in public.
Two things this skill is NOT:
- Not a deception tool. The point is closing the actual quality gap (the security holes and missing pieces ARE the tells), plus removing tool-attribution noise from what ships. It never fakes provenance, testimonials, stats, or history.
- Not a full redesign. The identity pass is bounded: token-level design changes and copy rewrites, not new layouts or features.
Workflow
Run as four phases: Audit → Report → Fix → Verify. Never skip the report — the user decides on the confirm-first items before you touch them.
Phase 1 — Audit (read-only)
Detect the stack first (package.json / pyproject.toml / framework config), then sweep every category in the two reference catalogs:
references/code-tells.md— security, reliability, code smells, missing pieces, repo/git tells. Each item has a concrete check (grep pattern or command). Run the checks; don't eyeball.references/design-tells.md— visual, copy, UX, and product-metadata tells, plus the identity playbook for Phase 3.
For repos over ~50 source files, fan the audit out to 2–3 parallel subagents (one for code/security, one for design/copy/metadata, one for repo/git hygiene) so the raw grep output stays out of the main context — each returns only its findings list. Set an explicit model on every subagent spawn.
Also run whatever the repo already has: linter, depcheck/vulture, npm audit, test suite. Existing tooling output counts as audit evidence.
What ships with it
3 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.
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.
- 11d ago First seen · 95 lines · 128 tokens per session scan A 6d5667582c5e
de-vibe is a skill published in the GitHub repository ucsandman/marketing-studio (234 stars, last pushed 4d ago), licensed MIT. It adds 128 tokens to every session and 1,898 once invoked, about $0.0006 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
open-design-landing-deck
OpenDesign's brand-story deck for partners and press: why it exists, what it believes, and where design is going. Built as a decision-grade marketing & GTM deck for partners, press, community.
html-ppt-zhangzara-block-frame
Rescuing a messy startup deck into a board-grade system — the diagnosis, the page grammar, and the rebuilt proof pages. Built as a decision-grade design craft deck for founders, exec presenter.
fs-emerald-editorial
OpenDesign's 'design on your desk' brand-launch narrative: the market moment, the story, and the proof that converts. Built as a decision-grade marketing & GTM deck for marketing team, press.
accesslint-scan
Audit a live page for accessibility issues, locate each WCAG violation precisely, and return a selector-grounded fix worklist without editing.
huashu-pentagram-grid
OpenDesign's positioning & messaging system: the one-line promise, the pillars, and the proof — the source of truth for all copy. Built as a decision-grade marketing & GTM deck for brand & marketing team.
check
Reviews code diffs, PRs, issue queues, release readiness, commits, pushes, publishing, and project audits. Use when users ask in any language for code review, issue or PR triage, release gates, publishing follow-through, or project audits. Not for debugging root causes or prose review.