Socialseal workspace setup
Skill OpenSealAI/agent-skills/skills/socialseal-workspace-setup
Use this skill when configuring a SocialSeal workspace for a brand, market, or campaign: selecting or creating the workspace, creating tracking groups, adding tracking items, running a baseline search journey, and confirming exports work. The deliverable is a working SocialSeal workspace, not a setup brief.From its SKILL.md
npx -y skills add OpenSealAI/agent-skills --skill socialseal-workspace-setupAssembled 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 file declares
Copied from the file, not written here
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
13.0 KB, ~2.9k tokens by cl100k_base, as published. Nobody here has run it
SocialSeal Workspace Setup
Overview
SocialSeal is a social-search intelligence platform. It helps teams understand what appears when people search on social and AI-search surfaces, and whether a brand, competitor, creator, or content pattern is discoverable for the right keywords.
A setup task is complete only when the workspace can produce usable data. A document is not the deliverable. The deliverable is a configured workspace with the right tracking groups, tracking items, baseline runs, and exports verified.
Core Concepts
| Concept | Meaning |
|---|---|
| Workspace | Top-level SocialSeal container for one brand, client, or project. Tracking groups and exports are scoped to a workspace. |
| Tracking group | A measurement container for one platform and one coherent keyword/topic scope. Example: TikTok / US / category searches. |
| Tracking item | A single item inside a group, usually a search keyword with region/platform metadata. |
| Search journey | A keyword-expansion or evidence run for a subject, subject type, and region. Useful for generating or validating keyword coverage before/after group creation. |
| Export | CSV/JSON output used by downstream analysis skills. Key export types include enriched search results and group evidence. |
When to Use
Use this skill for:
- New workspace onboarding.
- Selecting and validating an existing workspace.
- Creating tracking groups.
- Adding keyword tracking items to groups.
- Running baseline search journeys.
- Confirming that group evidence and enriched search-result exports work.
Do not use this skill for opportunity analysis, creator briefs, social plans, or measurement readouts. Those are downstream tasks that depend on a correctly configured workspace.
Inputs
Required
- SocialSeal CLI access or SocialSeal MCP server access.
- Workspace target: existing workspace ID/name, or enough context to identify the intended workspace.
- Brand or subject name.
- Target platform(s): use platform values supported by SocialSeal, commonly
tiktok,instagram,youtube,ig_reels,yt_shorts,douyin,xhs, orgoogle_ai. - Target market/region code, such as
US,GB,JP,SG,MY. - At least one keyword/topic set to track.
Good to have
- Owned social handles.
- Competitor handles or competitor brand names.
- Local-language keyword variants.
- Whether each group is branded, category, competitor, creator, or campaign tracking.
- Reporting cadence and expected downstream deliverable.
If the user is terse
If the user only says something like “set up SocialSeal for this brand,” ask for the minimum missing setup inputs:
- workspace or workspace name
- market/region
- platform(s)
- brand/subject
- seed keyword/topic list
If the workspace already exists and you can list workspaces, do that before asking.
Tooling
Both MCP and CLI work. Prefer MCP when available: discover with socialseal_list_workspaces and socialseal_list_available_tools, read socialseal_get_tool_schema before mutating calls, then socialseal_call_tool. The CLI mirrors this with tools list / tools schema / tools call. See references/mcp-and-cli-usage.md.
CLI Setup Workflow
Use npx -y @socialseal/cli unless the socialseal binary is already installed. Prefer machine-readable flags when available.
1. Discover the current CLI surface
Run these first. Do not rely on stale command memory.
npx -y @socialseal/cli --help
npx -y @socialseal/cli workspace --help
npx -y @socialseal/cli tools --help
npx -y @socialseal/cli data --help
Then inspect the relevant tool schemas:
npx -y @socialseal/cli tools list
npx -y @socialseal/cli tools schema --function group-management
npx -y @socialseal/cli tools schema --function search-journey-run
npx -y @socialseal/cli data export-options
2. Select the workspace
List accessible workspaces:
npx -y @socialseal/cli workspace list --pretty
Show the current default:
npx -y @socialseal/cli workspace current --pretty
Set the default workspace by ID, slug, or exact name:
npx -y @socialseal/cli workspace use <workspace-id-or-slug-or-exact-name>
If using scoped keys or automation, still pass --workspace-id <workspace-id> explicitly to later commands. This avoids accidental fallback to a personal/default workspace.
3. Design the group structure before creating groups
Use one group per platform and coherent measurement scope. A good group name encodes platform, market, and type.
Recommended group split:
TikTok / US / category searchesTikTok / US / branded searchesInstagram / US / category searchesYouTube / US / category searchesGoogle AI / US / category questions
Rules:
- Do not mix platforms in one group.
- Do not mix branded and category keywords in one group.
- Do not mix markets/languages unless the reporting question explicitly needs a combined view.
- Use local-language search phrases for non-English markets.
4. Create a tracking group
Supported platform values include tiktok, instagram, youtube, ig_reels, yt_shorts, douyin, xhs, and google_ai.
npx -y @socialseal/cli tools call \
--function group-management \
--workspace-id <workspace-id> \
--body '{"action":"create","name":"TikTok / US / category searches","platform":"tiktok","description":"Category search tracking for US TikTok keywords"}' \
--pretty
Save the returned numeric tracking group ID as <group-id>. Downstream export commands use numeric group IDs, not brand-group UUIDs.
5. Add tracking items to the group
For keyword payloads, use group_id and item objects. Omit item platform to inherit the group platform, or pass it explicitly if needed.
npx -y @socialseal/cli tools call \
--function group-management \
--workspace-id <workspace-id> \
--body '{"action":"add_items","group_id":<group-id>,"items":[{"name":"<keyword one>","type":"keyword","value":"<keyword one>","region":"US"},{"name":"<keyword two>","type":"keyword","value":"<keyword two>","region":"US"}]}' \
--pretty
Use real user search language, not internal marketing phrasing. For travel/hospitality and many consumer categories, useful keywords often express planning help, comparisons, timing, location, first-timer questions, or practical detail.
6. Check setup completeness
Use completeness to confirm expected memberships and optionally refresh visibility.
npx -y @socialseal/cli tools call \
--function group-management \
--workspace-id <workspace-id> \
--body '{"action":"completeness","group_id":<group-id>,"expected_items":[{"track_type":"search","track_value":"<keyword one>","region":"US"},{"track_type":"search","track_value":"<keyword two>","region":"US"}],"include_refresh_status":true}' \
--pretty
Do not declare setup complete until the group has the expected items.
7. Run or inspect a baseline search journey
A search journey helps validate subject, region, and keyword direction before or after group creation.
Start a run:
npx -y @socialseal/cli tools call \
--function search-journey-run \
--workspace-id <workspace-id> \
--body '{"subject":"<brand-or-topic>","subjectType":"brand","region":"US","executionMode":"async"}' \
--pretty
Poll an async journey run:
npx -y @socialseal/cli tools status <run-uuid> \
--kind journey_run \
--workspace-id <workspace-id>
Use subjectType: "brand" for a brand, "topic" for a category/topic. If the CLI schema shows additional fields such as locale, platformKeys, seedKeywords, contentPillars, or maxKeywords, use them when relevant.
8. Verify export data flow
List export options:
npx -y @socialseal/cli data export-options
For enriched ranked search rows:
npx -y @socialseal/cli data export-search-results \
--group-ids <group-id> \
--workspace-id <workspace-id> \
--out ./exports/search-results-<group-id>.csv \
--timeout 120000
For group evidence that automatically routes social and Google AI groups to the correct export shape:
npx -y @socialseal/cli data export-group-evidence \
--group-id <group-id> \
--workspace-id <workspace-id> \
--out ./exports/group-evidence-<group-id>.csv \
--timeout 120000
A valid setup should produce a file with usable rows or a clear, explainable reason for no rows, such as a new group with no completed refresh yet.
MCP Setup Workflow
Use the same logical workflow through MCP. Do not invent MCP tool names.
- List available MCP tools from the SocialSeal server.
- Identify tools corresponding to:
- workspace list/current/select
- tool/function schema or help
- tracking group creation
- tracking item add/bulk add
- group completeness/status
- search journey start/status
- search results or group evidence export
- Read tool schema before calling mutating actions.
- Create one group first, add a small keyword set, and verify completeness before bulk creation.
- Export evidence after setup to confirm the workspace produces downstream data.
If MCP lacks a setup operation that the CLI supports, use the CLI for that operation and continue with MCP for reads/exports.
Done Means
Workspace setup is done only when all of these are true:
- Correct workspace selected and confirmed.
- Planned tracking groups created with the correct platform values.
- Tracking items added to each group.
- Completeness check confirms expected items.
- At least one baseline search journey or status check has run for the workspace.
- At least one export command works for each important group type, or the reason for no rows is documented.
- Workspace ID, group IDs, group names, platform, market, and keyword set are recorded for downstream skills.
Do / Don't
Do
- Use
workspace listbefore setup if the workspace is not explicitly provided. - Use
tools schema --function group-managementbefore creating or adding items. - Use
group_id, notgroupId, ingroup-managementbodies unless the live schema says otherwise. - Use
--group-idsforexport-search-resultsand--group-idforexport-group-evidence. - Keep a raw setup log with commands run, returned IDs, and export file paths.
- Create and validate one group first, then scale the pattern.
Don't
- Don't call the output a brief when the user asked for setup.
- Don't mix platforms or keyword types inside one group.
- Don't add English-only keywords for non-English markets unless the user's scope is English-language search.
- Don't use brand-group UUIDs where the CLI expects numeric tracking group IDs.
- Don't treat a header-only export as success without explaining why no rows exist.
- Don't expose real workspace IDs, tracking IDs, API keys, or private exports in public examples.
Troubleshooting
Workspace not found
- Run
npx -y @socialseal/cli workspace list --pretty. - Confirm the key has access to the intended workspace.
- Use
workspace use <identifier>with ID, slug, or exact name. - Pass
--workspace-id <workspace-id>explicitly on scoped commands.
add_items succeeds but completeness fails
- Re-check the live schema for
group-management. - Confirm the body uses
group_idand item objects. - Confirm the region value is valid and consistent.
- Try adding a single item first to isolate malformed payloads.
Export returns only headers or no rows
- Confirm items were added with completeness.
- New or modified groups may need refresh time before ranked results exist.
- Try
export-group-evidenceto route the group to the correct export type. - Check date filters; too narrow a date range can hide valid results.
Search journey does not start
- Confirm required fields:
subject,subjectType,region, and workspace context. - Use
subjectType: "topic"for category searches, notbrand. - Add
executionMode: "async"for long runs and poll withtools status.
MCP and CLI disagree
- Prefer the live schema from the surface you are using.
- Record which surface created each group/item.
- If a mutating MCP call is missing or unstable, use the CLI for setup and MCP for inspection/exports.
Verification Checklist
- Correct workspace selected or identified.
- Group structure matches platform × market × keyword type plan.
- Groups created with valid platform values.
- Items added using schema-valid payloads.
- Completeness confirms expected tracking items.
- Baseline journey/status was run or intentionally skipped with reason.
- Exports tested and saved for downstream use.
- Setup log includes workspace ID, group IDs, group names, platforms, markets, and export paths.
- No private IDs or credentials appear in public/shared docs.
What ships with it: 1 file
7.8 KB alongside SKILL.md
references/
- mcp-and-cli-usage.md7.8 KB