agentleFS
sablier-labssablier-labs/agent-skills/skills/sablier-writing/SKILL.md

sablier-writing

Skill9 starsChanged 52 days agoMITView on GitHub

This skill should be used when writing content for Sablier, including "write a blog post", "create a case study", "draft a tweet", "write X/Twitter posts", "write an announcement", "create educational content", or any marketing content task for Sablier's brand.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create written content that represents the Sablier brand—informative, confident, and focused on real value rather than hype.

    line 8
  2. Distinctive:

    Before publishing any content:

    line 119
  3. Very specific:

    All references are located in the ./references/ directory:

    line 130

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Writing Skill

Create written content that represents the Sablier brand—informative, confident, and focused on real value rather than hype. This skill covers blog posts, case studies, and X/Twitter content.

Before Writing Any Content

Load shared brand references from the skill's references/ directory:

  1. Read ./references/brand-voice.md — Sablier's voice and style
  2. Read ./references/company-profile.md — products, metrics, positioning
  3. Read ./references/icp.md — who the audience is
  4. Reference ./references/voice-examples.md — past content for consistency
  5. Fetch docs if needed — use https://docs.sablier.com/llms.txt for product details

Content Types

Blog Posts

For long-form content: product announcements, educational articles, comparisons, technical deep dives.

Load detailed guidance: ./references/blog-posts.md

Quick reference for post types:

TypeLengthPurpose
Product Announcement800-1500wIntroduce new features/products
Educational1000-2500wTeach concepts, best practices
Case Study600-1000wCustomer success stories
Comparison1000-2000wCompare approaches/solutions
Technical Deep Dive1500-3000wDeveloper-focused technical docs

Case Studies

For customer success stories documenting how organizations use Sablier.

Load detailed guidance: ./references/case-studies.md

Standard structure:

  1. About [Customer] — 2-3 sentences
  2. The Challenge — what problem, why it mattered
  3. The Solution — why Sablier, how implemented
  4. The Results — quantifiable outcomes
  5. Customer Quote — direct quote if available
  6. Key Takeaway — one actionable insight

X/Twitter Posts

For social media content: announcements, educational threads, social proof, hot takes.

Load detailed guidance: ./references/x-twitter.md

Post type quick reference:

TypeStructure
AnnouncementNews → benefit → link
EducationalProblem → explanation → Sablier approach
Social ProofMetric → context → positioning
Case Study TeaserCustomer + outcome → quote → link
Hot TakePosition → reasoning → call to action

Voice Principles

Maintain these across all content:

  • Confident, not arrogant — Let data and track record speak
  • Educational first — Provide value even without Sablier
  • Direct — Get to the point, don't pad
  • Technical but accessible — Explain jargon, use examples
  • Customer-centric — Their story, not ours

Proof Points

Weave in when relevant:

  • 6+ years without hacks
  • 530,000+ streams created
  • Major customers: Ethena, Uniswap DAO, Fluid, Maple, Immutable
  • Low, predictable fees
  • Multiple audits
  • Operating since 2019

Good example:

Sablier has been running on mainnet since 2019 without any security incidents—a track record
that matters when you're distributing millions in tokens.

Bad example:

Sablier is the best and most secure platform!!!

What to Avoid

Across all content:

  • "We're excited to announce..." — Lead with the news
  • Hype words (revolutionary, game-changing, LFG)
  • Vague claims ("great results", "significant improvement")
  • Making it about Sablier instead of the reader/customer
  • Feature lists without benefits — Always include "so what"
  • Wall of text — Use headers, lists, short paragraphs
  • Engagement bait ("RT if you agree")
  • Excessive emoji

Quality Checklist

Before publishing any content:

  • Does it provide value even without Sablier mentions?
  • Are claims supported with specific data or examples?
  • Is it scannable (headers, lists, short paragraphs)?
  • Is voice consistent with brand guidelines?
  • Are CTAs clear and appropriate (not pushy)?
  • Does it sound like Sablier, not generic crypto marketing?

Reference Files

All references are located in the ./references/ directory:

Brand References

  • ./references/brand-voice.md — Voice and style guidelines
  • ./references/company-profile.md — Products, metrics, positioning
  • ./references/icp.md — Ideal customer profile
  • ./references/voice-examples.md — Past content examples

Content-Type Specific References

  • ./references/blog-posts.md — Detailed blog post templates and guidelines
  • ./references/case-studies.md — Case study structure and examples
  • ./references/x-twitter.md — X/Twitter post formats and examples

Load the appropriate content-type reference when writing that specific format.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.