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 inklate/social-skills --skill social-postgit clone --depth 1 https://github.com/inklate/social-skillsWrote 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/inklate/social-skills/social-post)<a href="https://agentmods.dev/skills/inklate/social-skills/social-post"><img src="https://agentmods.dev/badge/skills/inklate/social-skills/social-post/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/inklate/social-skills/social-post"><img src="https://agentmods.dev/badge/skills/inklate/social-skills/social-post.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 | $0.00176 | $0.01502 |
| Opus 5 | $0.00088 | $0.00751 |
| Sonnet 5 | $0.00035 | $0.00300 |
| Haiku 4.5 | $0.00018 | $0.00150 |
Grade A, and why
social-post 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 — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Write one platform-native post that sounds like the user and survives the feed: hook above the fold, one idea, room to breathe, a CTA that asks for something specific.
Context
Read social-context.md at the project root (also check .agents/social-context.md) for
Positioning, Voice rules, Audience, Pillars, Goals, and the Never list. If it's missing,
offer to run the social-context skill first, but don't block — ask three quick inline
questions (which platform, who's the audience, any hard nos) and proceed.
Workflow
- Confirm the platform and load its rules. One post targets one platform — ask which if
the user didn't say. Load references/platform-rules.md for
that platform's fold, length, link, hashtag, and formatting norms, and draft to them, not
from memory. (For the same message across several platforms at once, use
social-crosspost.) - Get the raw material immediately. Ask what the post is about if not stated; if the user gave a topic, dig for the specific: the number, the mistake, the exact moment, the line someone actually said. A post about "lessons from our launch" is dead; a post about "the day-3 metric that told us it was over" is alive. Don't draft until you have at least one concrete detail.
- Pick the angle. Map the material to one of the user's Pillars and check it against the Never list. If the material could serve two audiences (peers vs. buyers), ask which one this post is for — the same story is told differently to each.
- Ruthlessly cut to one idea. If the material contains two lessons, tell the user and ask which ships today (the other becomes a future post). A post that argues two things argues nothing.
- Draft the hook first, in two different patterns. The hook is everything above the fold —
the platform's visible-truncation point from the reference. Pick two contrasting patterns:
- Outcome-first: the result, no context. "We deleted our pricing page. Trials went up 40%."
- Confession: the mistake stated flat. "I spent $30k on a launch nobody noticed."
- Contrarian claim: the belief they're wrong about. "Your case studies aren't building trust. They're screening you out."
- Specific scene: drop into the moment. "Tuesday, 6:14am, our biggest customer's name in my inbox with the subject 'we need to talk'." Never open with throat-clearing ("I've been thinking a lot lately about…") and respect any Voice rule about openers.
- Write the body for each hook, sized to the platform: one thought per paragraph, a line break every 1–2 sentences where the platform allows it. Structure = tension → turn → takeaway. The takeaway must be something the reader can use, not a moral ("be resilient" fails; "kill the launch if week-1 activation is under 20%" passes).
- End with one specific CTA matched to the post's job and the user's primary goal (
## Goals):- a pointed question that has a real answer ("What's your kill threshold?"), or
- an offer ("I wrote up the full postmortem — comment 'launch' and I'll send it"), or
- a plain follow reason. Never "Thoughts?" or "Agree?".
- Place the link the platform's way (from the reference). Say why once when the platform penalizes body links; if the link belongs off-body, write the first-comment or bio version — one line of context plus the link.
- Apply the platform's hashtag norm from the reference: always specific (
#devtoolsnot#business), never a hashtag soup. - Self-check both variants against every Voice rule in
social-context.md, line by line. Fix violations; if a rule conflicts with a strong hook, keep the voice and note the tradeoff to the user. - Final pass — read each variant as it will render on the platform, then check both variants against every row of the Quality bar and the platform's rules in the reference. Fix any that fails. Count characters and show the hook and total counts.
- Present both variants labeled by hook pattern, with hook char count and total char count for each, plus the first-comment/bio link text if any. Ask which variant to refine, and iterate on the winner.
What ships with it
2 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 · 104 lines · 176 tokens per session scan A ff73dfe41111
social-post is a skill published in the GitHub repository inklate/social-skills (6 stars, last pushed 1mo ago), licensed MIT. It adds 176 tokens to every session and 1,502 once invoked, about $0.0009 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…