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-calendargit 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-calendar)<a href="https://agentmods.dev/skills/inklate/social-skills/social-calendar"><img src="https://agentmods.dev/badge/skills/inklate/social-skills/social-calendar/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-calendar"><img src="https://agentmods.dev/badge/skills/inklate/social-skills/social-calendar.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 | $0.00150 | $0.01741 |
| Opus 5 | $0.00075 | $0.00870 |
| Sonnet 5 | $0.00030 | $0.00348 |
| Haiku 4.5 | $0.00015 | $0.00174 |
Grade A, and why
social-calendar 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 12d 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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Build a realistic 2-week content plan from the user's pillars, platforms, and honest cadence — a plan, not a scheduler.
Context
Read social-context.md at the project root (also check .agents/social-context.md) for the ## Voice, ## Pillars, ## Platforms, ## Audience, ## Cadence, and ## Goals sections. If the file is missing, offer to run the social-context skill first, but don't block — ask 2–3 quick inline questions and proceed:
- What are your 2–4 content pillars?
- Which platforms, and which one matters most?
- How many posts per week have you actually sustained in the last month?
Workflow
- Confirm cadence reality before anything else. Take the number in
## Cadenceat face value — it's the user's honest count, so respect it. If the user asks for more volume than their stated cadence ("give me daily posts" when cadence says 3/week), push back once, explicitly: an unrealistic calendar is worse than none, because missed slots train the user to ignore the whole plan. Plan to the cadence they will actually keep, or to a new number they re-confirm out loud. - Fix the date range. Two weeks starting from the next natural posting day — usually tomorrow, or the coming Monday if today is late in the week. Ask about timezone only if an anchor date makes it genuinely ambiguous.
- Collect anchors. Ask whether anything real lands in the window: a launch, event, webinar, release, or a holiday their audience actually observes. Anchored content is placed first; the rest of the plan flows around it. Anything significant gets a small arc, not a single slot:
- one warm-up post 2–4 days before (tease the problem, not the announcement),
- the anchor-day post itself,
- one follow-through 1–3 days after (result, lesson, or behind-the-scenes).
- Distribute pillars. Spread pillars across the slots so no pillar appears twice in a row on the same platform, and no pillar exceeds ~40% of total slots unless the user says one pillar deliberately dominates. Weight the mix toward the primary goal from
## Goals— lead-gen tilts to pillars that show the product working; awareness and authority tilt to opinion and teaching. If they have 4+ pillars and few slots, don't shave every pillar to one token post — bench the weakest pillar for this fortnight and tell them which one and why. - Distribute formats. Rotate the formats each platform actually supports, and match format to pillar — teaching content suits threads, carousels, and document posts; opinion and story content suits single text posts:
- LinkedIn: text post, document/carousel post, image post
- X: single post, thread, image post
- Instagram: carousel, Reel idea, single image
- Threads / Bluesky: short text post, small thread
- Facebook: text post, image post, link post Never assign a format the platform doesn't support, and never let one platform run a single format for the whole fortnight.
- Write a working title or hook for every slot. Not a topic label — an actual first line the user could open with. "Why we killed our best feature" beats "product lessons". If the brand voice is in context, write hooks in that voice; if a Never-list exists, honor it here too.
- Reserve reactive slack. Leave roughly 20% of slots — at least 2 across the fortnight — labelled
reactivewith no assigned topic. These absorb trending conversations, a reply that deserves to become a post, and news in the user's niche. Say this explicitly in the plan notes so the user reads them as capacity, not gaps. - Sequence for energy, not neatness. Front-load week 1 with the strongest hooks — early wins keep the plan alive. Put heavy-prep formats (carousel, long thread) on days the user likely has time; ask if you don't know. If one idea suits two platforms, stagger it 1–3 days apart with a reworked hook — never a same-day copy-paste.
- Do a final realism pass. Before presenting, check the plan against every row of the Quality bar — volume, pillar spacing (including the ~40% cap), format mix, hooks, slack, anchors, and cross-posting stagger. Cut before you cram — a 9-slot plan the user completes beats a 14-slot plan they abandon.
- Offer, but don't produce unprompted, an .ics export. If the user wants calendar entries, generate a
.icsfile with one all-day event per slot titled[platform] hook, so the plan lands in whatever calendar they already look at.
What ships with it
1 file 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.
- 12d ago First seen · 86 lines · 150 tokens per session scan A 7c59d02fd242
social-calendar is a skill published in the GitHub repository inklate/social-skills (6 stars, last pushed 1mo ago), licensed MIT. It adds 150 tokens to every session and 1,741 once invoked, about $0.0007 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…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
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…