agentleFS
EditorialOSEditorialOS/pinterest-marketing-strategist/skills/client-context/SKILL.md

client-context

Skill2 starsChanged 7 months agoApache-2.0View on GitHub

Reads the client's Google Drive folder and image library to build structured Pinterest context. Extracts brand voice, audience, boards, image inventory with orientation analysis, and pin performance history. Called automatically at the start of every command.

3 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions3 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 3

    1. EditorialOS/editorial-os★ 0same name
    2. EditorialOS/email-marketing-manager★ 0same name
    3. itsKayumkhan/claude-client-context★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository2 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every /create and /track command starts here.

    line 18
  2. Distinctive:

    Read the ~~docs folder. Look for these in priority order:

    line 22
  3. Distinctive:

    After scanning, assess the library:

    line 74
  4. Distinctive:

    If any category has zero images but the pin queue includes topics that need that category, flag: "Pin queue includes [topic] but no …

    line 99

“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “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

Client Context — Pinterest Intelligence Layer

Purpose

Read the client's real documents and images to build the context layer that makes every composed pin client-specific. Pinterest is a visual search engine — context here means images and search terms as much as voice and audience. The image library inventory is the foundation of composed pin output.

Core Principle

This skill exists so that /create produces pins from real brand assets, not generic templates. Without context, pins are generic. Without images, pins are incomplete. Without learnings, pins don't improve. This skill loads all three.

When This Activates

Every /create and /track command starts here. Load context silently — show a one-line status header and move to the command's work. Don't narrate the loading process.

Priority Documents

Read the ~~docs folder. Look for these in priority order:

PriorityDocument TypeWhat to ExtractIf Missing
1Image libraryAvailable images — from ~~images (Cloudinary) or ~~docs images/ folder. Filenames, dimensions, orientation, categories, tags, visual descriptions. This is the most critical asset.Cannot produce composed pins. Flag: "No images found. Add brand images to create composed pins."
2Brand guide / style guideVisual identity, voice for descriptions, vocabulary rules, banned wordsAsk for brand name and 2-3 adjectives
3Product / service catalogNames, descriptions, prices, destination URLs, categoriesAsk what they sell and where it links
4Audience personasDemographics, interests, Pinterest search behavior, planning patternsAsk who the target Pinterest user is
5Board structureBoard names, descriptions, themes, goalsRecommend boards based on brand vertical
6Pin learning filespin-log.md, pin-learnings.md, pin-baselines.md from /trackFirst batch — no learning history
7Pinterest configpinterest-config.md from /setupRun /setup first or proceed with defaults
8Content strategyGoals, pillars, seasonal calendar, pin queueAsk for primary goal

Image Library Scanning

This is the most critical context step for composed pin production.

Source Priority

  1. ~~images (Cloudinary) — if connected, this is the primary image source. Advantages: automatic orientation data, metadata tags, transformation URLs for Pinterest sizing, CDN delivery.
  2. ~~docs images/ folder — if Cloudinary is not connected, scan the Drive images folder. Less metadata available but functional.
  3. Both — if both exist, Cloudinary is primary (better metadata and transforms), Drive is supplemental.

Image Inventory Protocol

For each image in the library, capture:

DimensionWhat to CaptureWhy It Matters
Filename / Asset IDExact reference for pin outputRequired for composed pin attribution
Dimensions (px)Width × heightDetermines orientation and Pinterest suitability
OrientationVertical (2:3), square (1:1), horizontal (3:2)Vertical gets 60% more engagement on Pinterest
Aspect ratioExact ratioDetermines if smart crop is needed for 2:3
CategoryProduct, lifestyle, team, diagram, quote card, seasonal, behind-scenesMatches pin concept type
Tags / metadataCloudinary tags or Drive folder conventionsEnables topic-based search during /create
Visual descriptionWhat's in the image — subject, setting, moodEnables concept-to-image matching
Quality signalsResolution adequate? Well-lit? Clear subject?Low-quality images hurt pin performance
Reuse statusUsed in last 10 pins? (from pin-log.md)Pinterest deprioritizes duplicate images
Transform URLCloudinary URL with Pinterest sizing appliedReady-to-use for composed pin output

Orientation Classification

OrientationAspect RatioPinterest PerformanceAction
Vertical2:3 or tallerOptimal — 60% more engagementUse as-is or crop to exactly 1000×1500
Square1:1Acceptable — less feed spaceCan use, or crop to vertical with smart crop
Horizontal3:2 or widerWorst performerSmart crop to vertical if possible, flag if not

Image Health Assessment

After scanning, assess the library:

ConditionStatusRecommendation
20+ vertical images🟢 HealthySufficient for sustained batches
10-19 vertical images🟡 ThinWill run low within 2-4 weeks of regular posting
1-9 vertical images🟠 CriticalAdd more vertical images before regular production
0 images🔴 EmptyCannot produce composed pins. Blocker.
All horizontal⚠️ MisfitPinterest penalizes horizontal. Add vertical shots.
No new images in 60+ days⚠️ StalePinterest rewards visual freshness. Add new photos.
Fewer images than batch size⚠️ InsufficientReduce batch size or add images

Category Distribution

Track the mix of image types:

CategoryIdeal ForCountStatus
Product shotsProduct spotlight pins, sales goal
LifestyleAspirational pins, save goal
Team / workspaceBehind-scenes, brand building
Diagrams / infographicsEducational, how-to pins
Quote cards / text overlayThought leadership
SeasonalTimely, event-driven pins

If any category has zero images but the pin queue includes topics that need that category, flag: "Pin queue includes [topic] but no [category] images exist. Add [specific suggestion] before creating pins for that topic."

Brand Voice (Adapted for Pinterest)

Pinterest copy is shorter and more search-oriented than email or blog. Extract voice differently:

DimensionWhere to Find ItPinterest Adaptation
Core visual identityBrand guideInforms image selection preferences and text overlay decisions
Description voiceBrand guide toneSame voice, fewer words. Pinterest descriptions are 150-300 chars.
Vocabulary — useWord list from guidePrioritize search-relevant terms from the list
Vocabulary — avoidBanned words from guideApply same restrictions in titles and descriptions
CTA styleBrand guide or past pinsPinterest CTAs are soft — "Tap to explore" not "BUY NOW"

Critical Pinterest voice rule: Descriptions serve search first, personality second. Front-load keywords. Save the brand character for the title. A SaaS brand's description: "Content workflow automation for stretched marketing teams. Reduce approval cycles by 60% with editorial infrastructure." Not: "We believe content should be joyful!"

Audience Segments (Pinterest-Specific)

Pinterest users behave differently from other channels:

DimensionWhat to Extract
Pinterest demographics60% female (male growing fastest), median age 25-44, high-income skew
Search behaviorWhat terms does this audience type into Pinterest?
Planning stagePlanning ahead (saves) or ready to act (clicks)?
Save vs. click profileSome audiences build boards. Some click through. Which is this?
Device85% mobile. All pins must look good at mobile dimensions.

Intelligence Levels

Data AvailableIntelligence Level
Personas onlyBasic — demographics, interests, likely search terms
Personas + pin historyModerate — knows which topics and image types engaged
Personas + history + learningsDeep — knows save rate by image-copy combo, CTR by board
All above + analyticsFull — real data per variable

Board Structure

For each configured board:

  • Name and description (descriptions should include keywords for Pinterest SEO)
  • Theme — what content belongs here
  • Goal — traffic, saves, awareness, sales
  • Pin count — from pin-log.md
  • Board health — getting fresh pins? Growing?
  • Best image-copy combo — from learning files (if tracked)

Board strategy rules:

  • Minimum 4 boards for serious presence
  • Each board needs 15-25 pins to look established
  • Board descriptions must include search keywords
  • Seasonal boards can be archived post-season but pins stay discoverable

Performance Baselines

SourceWhat to Read
pin-baselines.mdOverall averages, per-board, per-image-type, per-combo baselines
pin-learnings.mdCumulative learnings by category
pin-log.mdLast 20-30 pins with images used, boards, metrics

Confidence Scale

Tracked PinsConfidenceWhat the System Knows
0BaselinePinterest vertical benchmarks only
1-10LowDirectional. "Lifestyle images seem to save better."
11-30MediumBoard-level patterns. Image type preferences. Title effects.
31-75HighMulti-variable patterns. Image-copy combo predictions.
76+Very HighFull predictive capability. Specific combinations optimized.

Connector Awareness

ConnectorIf AvailableIf Unavailable
~~docsRead all documents. Write learning files back. Full context.Ask user to describe brand and paste info. No persistent learning.
~~imagesScan Cloudinary for image library. Get metadata, transforms, CDN URLs. Composed pins with ready-to-use images.Fall back to ~~docs images/ folder. Less metadata, no transforms.
~~pinterestPost pins via API. Pull analytics for /track.User posts manually. Pastes analytics for /track.
~~calendarSeasonal timing recommendations.Use general seasonal knowledge.

Degradation Behavior

LevelConnectedCapability
Full~~docs + ~~images + ~~pinterestComplete: context + composed images + auto-post + auto-track
Composed~~docs + ~~imagesStrong: context + composed images. User posts manually.
Docs only~~docs (with images/)Good: context + image matching from Drive. No transforms.
ManualNothingFunctional but limited: user describes brand, provides images. No learning persistence.

Status Header

Display one line at the start of every command:

  • 🟢 Config: Connected — [brand] — [N] images ([N] vertical) — [N] boards — [confidence level]
  • 🟡 Config: Partial — [what's found] — missing [what's not] — [image library status]
  • ⚪ Config: Manual — provide brand context and images

Cross-Plugin Compatibility

Same Drive folder serves Editorial OS, Email Marketing Strategist, and Pinterest Marketing Strategist. Shared: brand guide, audience personas, content strategy. Pinterest-specific: image library (shared with Composer), pin-log.md, pin-learnings.md, pin-baselines.md.

Teammate Protocol Compatibility

This skill works in both modes:

Plugin mode (Cowork): Reads from ~~docs and ~~images. User triggers /create and /track manually.

Teammate mode (Autonomous): Reads from workspace Drive folder. Cron triggers create and track per standing orders. Same skill, different trigger. Same learning files, same image library.

Standing orders example:
- Task: Create composed pin batch
  Day: Monday, Wednesday, Friday
  Skill: pin-strategist + client-context
  Image source: Cloudinary folder "editorial-os/pinterest"
- Task: Track pin performance
  Day: Monday
  Skill: performance-learning

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.