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/brand-voice-keepernpx skills add moses607/socialforge --skill brand-voice-keepergit clone --depth 1 https://github.com/moses607/socialforgeWrote 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/moses607/socialforge/brand-voice-keeper)<a href="https://agentmods.dev/skills/moses607/socialforge/brand-voice-keeper"><img src="https://agentmods.dev/badge/skills/moses607/socialforge/brand-voice-keeper.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 | $0.00107 | $0.00939 |
| Opus 5 | $0.00053 | $0.00469 |
| Sonnet 5 | $0.00021 | $0.00188 |
| Haiku 4.5 | $0.00011 | $0.00094 |
Grade A, and why
brand-voice-keeper 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 — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Brand Voice Keeper
A brand is recognized before it is read — the same voice and look across dozens of posts is what turns scattered content into a memorable presence. Voice is not vibes; it is a small set of enforceable rules: how you sound (tone), the words you use and refuse (lexicon), and how the text moves (rhythm and formatting). Define it once from real examples, then gate every draft against it. Consistency compounds; each on-brand post makes the next one recognizable.
1. Distill the voice
- Pick 3-5 tone attributes as tension pairs, e.g. "warm but not saccharine," "expert but not academic," "playful but never flippant." Vague single words (friendly, professional) fail — force the "not."
- Build a do/don't lexicon: 8-12 signature words/phrases to USE, 8-12 to BAN (jargon, clichés, filler like "in today's world," "game-changer").
- Set rhythm rules: sentence length (short punchy vs. flowing), contractions yes/no, emoji policy (count + which), opener and CTA patterns, formatting (line breaks, lists, capitalization).
2. Extract voice from samples
- Paste 3-5 of the user's best-performing or favorite posts.
- Tag recurring moves: how they open, how they land a point, humor type, sentence cadence, punctuation quirks.
- Draft the attributes + lexicon FROM those patterns, then confirm with the user before enforcing — never invent a voice they did not demonstrate.
3. Enforce with the checklist
- Run every draft line-by-line against the Voice Guide below.
- Mark each check PASS or FAIL with the offending snippet quoted.
- Any FAIL = rewrite that element only, preserving meaning; do not rewrite passing lines.
Output template
VOICE GUIDE
Tone attributes: <A but not X> | <B but not Y> | <C but not Z>
Use words: <8-12 words/phrases>
Ban words: <8-12 words/phrases>
Rhythm: sentences <short/mixed> | contractions <yes/no> | emoji <n + which> | opener <pattern> | CTA <pattern>
Visual: colors <hex primary/secondary/accent> | fonts <heading/body> | framing <thumbnail/cover style> | logo <placement>
CONSISTENCY CHECK — <draft title>
[ ] Tone matches all attributes ......... PASS / FAIL: "<snippet>"
[ ] No banned words .................... PASS / FAIL: "<snippet>"
[ ] Rhythm + formatting on-brand ....... PASS / FAIL: "<snippet>"
[ ] Opener + CTA follow pattern ........ PASS / FAIL: "<snippet>"
[ ] Emoji/punctuation within policy .... PASS / FAIL: "<snippet>"
[ ] Visual assets on-brand ............. PASS / FAIL: "<snippet>"
VERDICT: ON-BRAND / NEEDS FIX
Fixes: <only the failing elements, rewritten>
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 · 55 lines · 107 tokens per session scan A 82f87054478e
brand-voice-keeper is a skill published in the GitHub repository moses607/socialforge (2 stars, last pushed 1mo ago), licensed MIT. It adds 107 tokens to every session and 939 once invoked, about $0.0005 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
self-media-content-workflow
通用自媒体内容生产与经营工作流。用于自媒体内容创作、选题策划、热点或竞品研究、多平台改写、短视频脚本、数字人视频、发布包、公众号草稿、内容数据分析、周复盘、月复盘和继续未完成任务。负责识别请求类型，调用创作简报、内容策略、热点竞品、平台文案、短视频、数据复盘和交付归档模块，并管理方向确认、标题确认、终稿确认和发布授权。.
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…
smart-illustrator
智能配图与 PPT 信息图生成器。支持三种模式：(1) 文章配图模式 - 分析文章内容，生成插图；(2) PPT/Slides 模式 - 生成批量信息图；(3) Cover 模式 - 生成封面图。所有模式默认生成图片，--prompt-only 只输出 prompt。支持 Bento Grid 功能展示图风格（--style bento）。触发词：配图、插图、PPT、slides、封面图、thumbnail、cover、bento grid、功能展示图、feature showcase。.