agentsclimarketplace

Socialseal orchestrator

Skill OpenSealAI/agent-skills/skills/socialseal-orchestrator

Open-source Agent Skills for SocialSeal: social-search strategy, the vNext creator/UGC production engine (reference videos -> Video DNA -> blueprint -> brief -> Asset Studio), and discoverability measurement.

Install
npx -y skills add OpenSealAI/agent-skills --skill socialseal-orchestrator

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use this skill when starting, scoping, or unsure how to approach any SocialSeal task (strategy, setup, analysis, content production, or measurement). It is the lightweight entry point: it checks foundations first, routes to the right SocialSeal skill in the right order, and keeps the agent proactive instead of waiting for a perfectly specified request.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

5.9 KB, ~1.2k tokens by cl100k_base, as published. Nobody here has run it

SocialSeal Orchestrator

Overview

This is the always-on router for SocialSeal work. Keep it light: its job is to figure out where the user is, check that foundations exist, and hand off to the right skill. It does not do the deep work itself.

SocialSeal concepts (personas, pillars, discoverability, share of voice, tracking groups, blueprints) are not widely understood. Be proactive: diagnose what is missing, explain it briefly, and offer to produce it with SocialSeal, rather than asking the user to supply perfect inputs.

First move: choose access mode and check foundations

Before strategy or setup checks, confirm whether live SocialSeal tools are available:

  • If socialseal_list_workspaces, socialseal_list_available_tools, and socialseal_get_tool_schema are available, use MCP mode. Start with workspace discovery and live registry/schema inspection.
  • If no socialseal_* tools are available in Cowork or another non-technical environment, explain that the hosted SocialSeal connector is not connected or enabled. Tell the user to open Customize -> Connectors, click + -> Add custom connector, fill in Name socialseal and Remote MCP server URL https://mcp.socialseal.co/mcp, click Add/Connect, sign in, then retry.
  • If the connector is unavailable, switch to file mode. Ask for SocialSeal CSV/JSON exports and continue with the skills that can work from files.
  • If the user is in Claude Code or a developer environment, local stdio MCP and CLI are fallbacks. They require Node.js/npx and must be installed separately.

See references/onboarding-and-auth.md and references/mcp-and-cli-usage.md. Never print raw ss_cli_... keys in full; use only the final six characters.

Before any analysis or production, decide whether strategy and setup exist. If you are unsure, run socialseal-strategy-readiness first; it diagnoses strategy foundations (personas, pillars, brand voice, goals) and SocialSeal setup, and guides the user to define what is missing using SocialSeal research.

Do not stall on a missing input. Name it, explain it in a sentence, propose a SocialSeal-backed way to define it, then route.

Routing map

  • Missing socialseal_* tools in Cowork -> hosted connector setup from references/onboarding-and-auth.md, then retry workspace discovery
  • Connector unavailable -> file mode with user-provided SocialSeal exports, then route to the best analysis skill
  • Claude Code developer fallback -> local stdio MCP or CLI setup from references/onboarding-and-auth.md, then resume
  • Invalid local MCP/CLI credentials -> device login from references/onboarding-and-auth.md, then resume
  • Unclear request, or strategy/setup may be missing -> socialseal-strategy-readiness
  • Need a workspace / tracking groups created or fixed -> socialseal-workspace-setup, socialseal-tracking-group-design
  • "Where should we make content?" / gaps -> socialseal-opportunity-analysis
  • "What are competitors/creators doing?" -> socialseal-competitor-content-analysis
  • "Which creators should we partner with?" -> socialseal-creator-discovery
  • "What's trending in local language vs English?" / catch micro-trends early -> socialseal-bilingual-demand-monitoring
  • "Where is demand shifting?" / route budget and fast-track tour onboarding -> socialseal-predictive-demand-routing
  • Turn opportunities into a plan -> socialseal-social-plan-builder
  • Produce content (engine path): socialseal-video-concepting -> socialseal-reference-video-analysis -> socialseal-blueprint-builder -> socialseal-creator-briefing -> socialseal-asset-planning / socialseal-generation-prompts -> socialseal-asset-studio-generation -> socialseal-capcut-export-prep
  • "How did it do?" / movement -> socialseal-performance-readout, socialseal-discoverability-tracking
  • Decide next changes -> socialseal-content-adjustment-recommendations
  • Brief leadership -> socialseal-management-reporting
  • Turn a meeting/report into actions -> socialseal-follow-up-planning

Workflow order (evidence-first)

Strategy & setup -> opportunity/competitor analysis -> plan -> production engine -> measurement -> recommendations/reporting -> follow-up. Production is grounded in real surfacing videos via the blueprint engine; do not jump to "write a content idea" when the engine exists. See references/production-pipeline.md.

Hard rules to carry into every routed task

  • Cite evidence in human-readable terms: "keyword" [market, platform], video title/URL, @handle, group name. Keep video_uid/search_result_id as internal traceability only. See references/socialseal-data-contract.md.
  • Be honest about evidence tiers: hard measurements are exact (not estimates), statistics carry selection bias (only high-ranking videos for tracked queries), and creative exemplars are anecdotal, not proof. See references/evidence-and-confidence.md.
  • Reuse one opportunityKey across blueprint, brief, and asset.
  • Never expose literal ids/tokens in shared artifacts; use placeholders.

Do / Don't

Do:

  • diagnose foundations before deep work and route accordingly
  • stay proactive: teach the concept and offer to produce it
  • hand off to one skill at a time and keep scope clear

Don't:

  • do the deep analysis/production inside this skill
  • wait for a perfect brief when you can guide the user to one
  • proceed with analysis when strategy/setup is clearly missing

What ships with it: 5 files

28.6 KB alongside SKILL.md

references/

Gives 0 of the 12 instructions most agent orchestration skills give in ~1.2k tokens

Counted across 742 of the 995 authors here whose files we hold, read 2026-08-07

  • Reference existing artifacts by path or URLin 53 of 742, across 25 files
  • Run the full test suite after integrating changesin 51 of 742, across 19 files
  • Dispatch one agent per independent problem domainin 50 of 742, across 17 files
  • Verify fixes do not conflictin 45 of 742, across 13 files
  • Include a suggested skills section in the documentin 45 of 742, across 17 files
  • Redact sensitive informationin 41 of 742, across 11 files
  • Save to the temporary directory of the operating systemin 39 of 742, across 10 files
  • Tailor the document to user-provided focus argumentsin 39 of 742, across 9 files
  • Spot check agent changes for systematic errorsin 34 of 742, across 7 files
  • Write a handoff document summarising the current conversationin 31 of 742, across 6 files
  • Assign each agent a specific scopein 23 of 742, across 8 files
  • Provide specific scope and clear goalin 23 of 742, across 5 files

Said here and by no other author read

  • check foundations before doing deep work
  • diagnose missing inputs and explain them briefly
  • run strategy readiness if foundations are missing
  • hand off to one skill at a time
  • keep scope clear during handoff
  • route to appropriate skills in correct workflow order

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.