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 skills/moses607/socialforge/analytics-interpreternpx skills add moses607/socialforge --skill analytics-interpretergit clone --depth 1 https://github.com/moses607/socialforgeWhat 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 | $0.00082 | $0.01128 |
| Opus 5 | $0.00041 | $0.00564 |
| Sonnet 5 | $0.00016 | $0.00226 |
| Haiku 4.5 | $0.00008 | $0.00113 |
Grade A, and why
analytics-interpreter 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 yesterday.
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 — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Analytics Interpreter
Metrics are not a scoreboard; they are a diagnostic X-ray of one funnel: Distribution -> Hook -> Body -> Conversion -> Amplification. Every number is evidence about exactly one stage. Growth stalls because ONE stage leaks, not because "everything is bad." Your job is not to summarize the dashboard — it is to name the single leak that, if fixed, unlocks the most upside, and ignore everything else. Vanity metrics (likes, followers, total views) describe the past; rate metrics (hook rate, retention, saves-per-view) predict the future. Diagnose rates.
1. Map each metric to what it REVEALS
- Impressions / reach -> DISTRIBUTION. How many the algorithm tested you on. Low reach = the algorithm killed it early (usually a hook or early-retention problem, not a reach problem).
- Hook rate / 3s-view rate (views ÷ impressions) -> HOOK QUALITY. Below ~30% weak, 30-45% average, 45%+ strong. This is the first gate.
- Average watch time & retention curve -> BODY/CONTENT QUALITY. For short video, watch-time ratio (avg watch ÷ length) above ~0.8 is strong; full watch or rewatch (>1.0) triggers pushes.
- CTR (on titles/thumbnails, YouTube/blogs) -> PACKAGING. 2-4% baseline, 5%+ strong, sub-2% weak.
- Saves & shares -> VALUE + IDENTITY. THE growth signals. Save = "useful to future me." Share = "this represents me." Target saves+shares ≥ 1-2% of views.
- Follows-per-view -> PROFILE + CONTENT FIT. Are viewers converting to subscribers.
- Comments -> RESONANCE. Emotional or debate-worthy enough to react.
2. Read the retention curve — the drop tells you what to fix
- Cliff in first 1-3s -> hook fails / mismatch between hook promise and thumbnail-or-first-frame. Fix the opening.
- Steady slow decline -> normal; healthy content loses viewers gradually. Leave it.
- Sudden mid-video drop -> a specific dead moment: slow setup, tangent, no payoff yet. Cut it.
- Flat / rising line -> loops, open loops, or payoff pulling viewers through. Do MORE of this.
- Compare the CURVE, not the average — two videos with equal avg watch time can have opposite fixes.
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.
- yesterday First seen · 64 lines · 82 tokens per session scan A e5832247df7b
analytics-interpreter is a skill published in the GitHub repository moses607/socialforge (2 stars, last pushed 1mo ago), licensed MIT. It adds 82 tokens to every session and 1,128 once invoked, about $0.0004 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 skills, from other repositories
blog-schema
Generate complete JSON-LD schema markup for blog posts with Article/BlogPosting, Person, Organization, BreadcrumbList, ImageObject, and optional FAQPage. Validates against Google requirements and warns about deprecated types. Use when user says "schema", "blog schema", "json-ld", "structured data", "schema markup"…
blog-audio
Generate audio narration of blog posts using Google Gemini TTS. Supports summary narration, full article read-aloud, and two-speaker podcast/dialogue mode with 30 voice options. Outputs MP3 with HTML5 audio embed code. Works standalone via /blog audio or internally from blog-write. Falls back gracefully when API key…
self-media-short-video
把已确认的母题或文案转成可直接拍摄、录屏或交给视频工具制作的短视频方案，也支持在用户确认肖像与声音权利并完成平台手动上传后制作数字人视频。用于视频号、抖音、小红书视频和其他竖屏短视频的口播稿、前 3 秒钩子、分镜、字幕、录屏清单、封面、话题、BGM 建议、数字人制片包和发布文案。.
self-media-content-delivery
将自媒体创作里程碑保存成文件，管理版本、核验路径、生成完整发布包并更新内容索引。用于保存平台第一版完整成稿、多平台合集、短视频脚本、分镜、定稿、发布包和复盘报告，或用户说“保存一下、整理交付、归档、生成发布清单”时。标题微调、删一句和临时候选不单独保存新版本。.
self-media-trend-radar
安全追踪热点、研究关键词和拆解竞品内容，并将结果转成原创选题和证据包。用于用户要求找热点、看趋势、分析爆款、研究竞品账号、拆标题或开头、比较内容结构、发现评论需求和判断时效窗口。只做公开信息研究和只读采集，不自动互动、发布或使用主账号登录态抓取竞品。.
x-marketing
Plan, draft, audit, and publish posts and threads for X (Twitter). Use when the user wants to write a single tweet or an auto-numbered thread, build a long-form tweetstorm, remove AI tells from a draft, reverse-engineer the hook from a viral tweet, draft a reply or quote tweet, or plan a week of X content. Tweets and…