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 surfmind-space/awesome-surfmind --skill caption-assistantgit clone --depth 1 https://github.com/surfmind-space/awesome-surfmindWrote 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/surfmind-space/awesome-surfmind/caption-assistant)<a href="https://agentmods.dev/skills/surfmind-space/awesome-surfmind/caption-assistant"><img src="https://agentmods.dev/badge/skills/surfmind-space/awesome-surfmind/caption-assistant/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/surfmind-space/awesome-surfmind/caption-assistant"><img src="https://agentmods.dev/badge/skills/surfmind-space/awesome-surfmind/caption-assistant.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00063 | $0.00561 |
| Opus 5.5 | $0.00025 | $0.00224 |
| Sonnet 5.5 | $0.00013 | $0.00112 |
| Haiku 4.5 | $0.00006 | $0.00056 |
Grade A, and why
caption-assistant 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 1mo 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 — 32 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Caption Assistant
Writes captions, accessibility alt text, and short descriptions for the images or content on the page, tailoring each to its purpose — with meaningful, fluff-free alt text for accessibility especially.
- Identify the target channel from the request and context: choose alt text when the user mentions screen readers, a11y, or accessibility; a social caption when they name a platform (Instagram, X, LinkedIn) or want engagement; a product caption or metadata for catalog and SEO; a short description otherwise. When unclear, default to alt text plus a short caption only.
- Describe the meaningful visual or page information accurately and concisely, drawing on the selection, visible page, attached context, and any tool results.
- For alt text, prioritize accessibility over marketing tone. Keep it under ~125 characters. For actionable images like icons in links or buttons, describe the action or destination in the image's own alt text (for example,
alt="Add to cart"), unless adjacent text already supplies that accessible name and the image itself is not being asked for. Use empty alt text (alt="") only for purely decorative images. - For social captions, include hook, context, and call to action only when the user asks for a social/platform caption or engagement copy, and respect the platform's length limits.
Lead with the alt text, then a short caption, then optional social caption variants only when they fit the request. Be specific without overclaiming, preserve important names, numbers, links, and user constraints, and do not identify private people or sensitive attributes unless provided and necessary. Do not describe non-visible details as if observed in any alt text, caption, social variant, product copy, or note. If the visible context is insufficient, say what is missing and give the best supported next step.
Example
Before (request: "caption this product photo for Instagram" — a person holding a steaming ceramic mug at a sunny kitchen counter):
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.
- 1mo ago First seen · 32 lines · 63 tokens per session scan A ee8aa02b52a0
caption-assistant is a skill published in the GitHub repository surfmind-space/awesome-surfmind (7 stars, last pushed 8d ago), licensed MIT. It adds 63 tokens to every session and 561 once invoked, about $0.0003 per session on Opus 5.5. 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 skills, from other repositories
cs-master
Expert computer scientist for algorithms, complexity analysis, and distributed systems.
uiux-master
UI/UX expert for design systems, accessibility, and avoiding generic AI-generated design patterns.
c-master
C programming expert for memory management, systems programming, and performance optimization.
data-science
ML engineering expert for feature engineering, model debugging, and production pipelines.
go-master
Go expert for goroutines, channels, microservices, and concurrent patterns.
network-devops
Network performance expert for latency optimization, packet analysis, and infrastructure tuning.