Mindsdb MCP Skill
LeoYeAI/openclaw-master-skills
MindsDB MCP服务器交互技能，用于通过自然语言查询和操作200+企业级数据源。当用户需要查询数据库、分析数据、创建AI模型、连接数据源（MySQL、PostgreSQL、MongoDB、Excel、CSV、Gmail、Slack等）、执行SQL查询、进行数据预测、构建知识库（RAG）、智能问答、文档检索或任何与数据库交互的任务时使用此技能。即使没有明确提到MindsDB，只要涉及数据库操…
Build and edit Tableau Next semantic models (SDMs), vizzes, and dashboards via MCP.
Install Tableau Next Asset Create
$ npx skills add forcedotcom/sf-skills --skill tableau-next-asset-create -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install forcedotcom/sf-skills tableau-next-asset-create --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/forcedotcom/sf-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/tableau-next-asset-create .claude/skills/tableau-next-asset-create && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "tableau-next-asset-create" agent skill from https://github.com/forcedotcom/sf-skills/tree/main/skills/tableau-next-asset-create into .claude/skills/tableau-next-asset-create/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "tableau-next-asset-create", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/forcedotcom/sf-skills/tree/main/skills/tableau-next-asset-createType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add forcedotcom/sf-skills --skill tableau-next-asset-create -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install forcedotcom/sf-skills tableau-next-asset-create --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/forcedotcom/sf-skills.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/tableau-next-asset-create .agents/skills/tableau-next-asset-create && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "tableau-next-asset-create" agent skill from https://github.com/forcedotcom/sf-skills/tree/main/skills/tableau-next-asset-create into .agents/skills/tableau-next-asset-create/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "tableau-next-asset-create", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add forcedotcom/sf-skills --skill tableau-next-asset-create -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install forcedotcom/sf-skills tableau-next-asset-create --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/forcedotcom/sf-skills.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/tableau-next-asset-create .cursor/skills/tableau-next-asset-create && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "tableau-next-asset-create" agent skill from https://github.com/forcedotcom/sf-skills/tree/main/skills/tableau-next-asset-create into .cursor/skills/tableau-next-asset-create/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "tableau-next-asset-create", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/forcedotcom/sf-skills.git --path skills/tableau-next-asset-create--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add forcedotcom/sf-skills --skill tableau-next-asset-create -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install forcedotcom/sf-skills tableau-next-asset-create --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/forcedotcom/sf-skills.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/tableau-next-asset-create .gemini/skills/tableau-next-asset-create && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "tableau-next-asset-create" agent skill from https://github.com/forcedotcom/sf-skills/tree/main/skills/tableau-next-asset-create into .gemini/skills/tableau-next-asset-create/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "tableau-next-asset-create", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install forcedotcom/sf-skills tableau-next-asset-createInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add forcedotcom/sf-skills --skill tableau-next-asset-create -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/forcedotcom/sf-skills.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/tableau-next-asset-create .github/skills/tableau-next-asset-create && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "tableau-next-asset-create" agent skill from https://github.com/forcedotcom/sf-skills/tree/main/skills/tableau-next-asset-create into .github/skills/tableau-next-asset-create/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "tableau-next-asset-create", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add forcedotcom/sf-skills --skill tableau-next-asset-create -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install forcedotcom/sf-skills tableau-next-asset-create --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/forcedotcom/sf-skills.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/tableau-next-asset-create .opencode/skills/tableau-next-asset-create && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "tableau-next-asset-create" agent skill from https://github.com/forcedotcom/sf-skills/tree/main/skills/tableau-next-asset-create into .opencode/skills/tableau-next-asset-create/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "tableau-next-asset-create", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
tableau-next-asset-create- GitHub stars
- 1.1k
- Token cost
- ~6.1k tokens
- SKILL.md length
- 2,289 words
- Files
- 65 (incl. references)
- Skills in repo
- 251
- Repo updated
- First seen
- Licence
- Apache-2.0
At a glance
Build and edit Tableau Next semantic models (SDMs), vizzes, and dashboards via MCP.
- : building a semantic model
- SKILL.md covers G1 — Precondition (assert…, Dispatch — route your intent…, No-dispatch rule (decline — do… and Deep references (cited by task…
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
- Dashboard from CSV/Excel files
What it does
Tableau Next Asset Create is an agent skill from forcedotcom/sf-skills. Build and edit Tableau Next semantic models (SDMs), vizzes, and dashboards via MCP. TRIGGER when: building a semantic model or dashboard from CSV/Excel files; ingesting a flat file/database into DLO/DMO; profiling a dataset (grain, measures, dimensions; large/wide files via inferobjectschema); adding a dimension, measure, AI-ready metric, join, or logical view; running a semantic query, interpreting a breakdown, analyzing account/customer engagement, or pulling counts/totals from existing Tableau Next data (even…
Its SKILL.md is about 6.1k tokens, which your agent loads only when the skill is triggered. The skill folder holds 66 other files, including reference files (for example `README.md`, `references/admin-users.md` and `references/ai-readiness.md`).
It sits in Data & Analytics, covering CSV and tabular files, Excel spreadsheets and SQL. It works with Tableau, Model Context Protocol, SQL and Microsoft Excel. The repository describes itself as: Salesforce's curated collection of agent skills for building applications. Optimized for Agentforce Vibes, compatible with all AI tools. The licence is Apache-2.0.
When your agent uses it
- : building a semantic model
- Dashboard from CSV/Excel files
- Ingesting a flat file/database into DLO/DMO
- Profiling a dataset (grain
Example prompts
- “alert me when”
- “/tableau-next-asset-create”
What it can do on your machine
Read from SKILL.md and the folder at commit e5164d9. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Tableau Next Asset Create loads about 6.1k tokens when it runs, and up to ~140k if it reads all its reference files. Until then it costs about 256 tokens; SKILL.md has 2,289 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from forcedotcom/sf-skills at commit e5164d9, republished under its Apache-2.0 licence (© forcedotcom). 2,289 words, ~6,070 tokens.
Download SKILL.mdSave it as
.claude/skills/tableau-next-asset-create/SKILL.md (or your agent's skills folder). This skill also uses 64 other files; get the full folder from GitHub.- name
- tableau-next-asset-create
- description
- Build and edit Tableau Next semantic models (SDMs), vizzes, and dashboards via MCP. TRIGGER when: building a semantic model or dashboard from CSV/Excel files; ingesting a flat file/database into DLO/DMO; profiling a dataset (grain, measures, dimensions; large/wide files via `infer_object_schema`); adding a dimension, measure, AI-ready metric, join, or logical view; running a semantic query, interpreting a breakdown, analyzing account/customer engagement, or pulling counts/totals from existing Tableau Next data (even unnamed/read-only); creating/editing a viz (filter, forecast, sort, shelf, mark encoding, reference line), widget, or dashboard layout. Also: Data Cloud SQL transforms; show a dashboard; share/revoke access; reuse prod or promote Personal Org assets; provision users/licenses; data alerts ('alert me when'); list/classify verified questions or score agent SQL. DO NOT TRIGGER for: rewriting prose descriptions, layout-only dashboard review, debugging MCP, or writing app code.
- metadata.version
- 1.1
- metadata.domains
- Tableau Next
Tableau Next SDM Authoring — Router
Build semantic data models that actually return data, then enrich, query, and visualize them. This skill encodes the failures that produce empty or confidently-wrong dashboards in Tableau Next — building on empty source objects, leaving objects unjoined, inventing field apiNames, summing an ID, inflating a snapshot by its date count — and carries through to enriching, querying, and visualizing the model.
This file is a router. It does two things: assert the one universal
precondition (G1), then dispatch your intent to the right task guide. The
guardrails that overlap between steps live in exactly one place —
references/shared-gates.md (gates G1–G8) — and each task guide opens with
the gates it must assert. Read the gate file once; every guide points back to it.
The MCP tools are exposed under a tableau-next-* server (Claude Code, for
example, advertises them as mcp__tableau-next-pilot-production__add_semantic_model_relationship).
Every tool name in this skill — router and reference/task files alike — is written
bare (e.g. add_semantic_model_relationship) for readability; invoke each one
under whatever name your client exposes for that server's tool. Match the advertised
name — do not hand-construct it. (Single statement: references/shared-gates.md.)
G1 — Precondition (assert once, here, before dispatching)
Every step of this skill drives a Tableau Next MCP server; without one connected
there is nothing to call. Before dispatching to any guide, confirm the available
tools include one prefixed mcp__tableau-next-*__ (e.g.
mcp__tableau-next-pilot-production__browse_data_assets,
mcp__tableau-next-self-service__create_semantic_model). Server names vary — match
the tableau-next stem, not an exact name.
If NO such tool is present, STOP. Do not attempt the workflow, do not describe
payloads as if you had run them, and never fabricate tool calls, results, or IDs.
Tell the user the skill requires a connected Tableau Next MCP server (e.g.
tableau-next-pilot-*, tableau-next-self-service) and ask them to connect one
and re-invoke. Every entry point inherits this — a standalone "add a metric" with
no server must still STOP. Full statement: references/shared-gates.md G1.
Dispatch — route your intent to a task guide
Classify on what the user wants to do, then read that guide. Each guide's
header names the gates it asserts (from references/shared-gates.md).
Question phrasing does not exempt an intent from classification. "How do I
profile X" → profile-dataset.md. "What payload do I use for
create_visualization" / "which chart type for a stacked bar" →
create-viz.md (not viz-authoring.md — that file is cited by the
guide, it is not a first hop). "Before I build, how should I design the
layout" → dashboard-design-principles.md (load this skill; do not answer
from general dashboard knowledge). Only an explicit disclaimer of
authoring ("don't build/create anything") sends advice / review / ideation
to the No-dispatch rule. A disclaimer that still asks for a number, table,
or interpretation ("don't build or change any model — just get the count
from existing Tableau Next data") is a query/analyze intent: route to
query-model.md / analyze-data.md. The question mark is never the test.
| User intent | Route to |
|---|---|
| "analyze this CSV" / "build a dashboard from these files" / nothing modeled yet | references/tasks/build-end-to-end.md |
| "ingest this CSV / Excel into a DLO/DMO" or ingest from an existing Snowflake / Databricks / BigQuery / Redshift connection (ingest only, no downstream ask) | references/tasks/ingest-flat-file.md |
| "create / update a data transform" / derived DLO or DMO via Data Cloud SQL | references/tasks/create-data-transform.md |
| "profile / discover / check this data" (read-only) | references/tasks/profile-dataset.md |
| add a raw object / dimension / measure / join / logical view / parameter to my model | references/tasks/edit-sdm.md |
| add a calculated dimension / measure / metric to my model | references/tasks/enrich-model.md |
| "is my model AI-ready" / "backfill readiness metadata" / validate a model | references/tasks/ai-readiness-audit.md |
count / total / scalar ("how many orders", "what is total revenue") or named-model structured run_semantic_query or raw DLO/DMO run_query | references/tasks/query-model.md |
| interpreted breakdown / ranking / trend / comparison without asking to author | references/tasks/analyze-data.md |
| delete a calc / metric / data object / logical view / relationship (component, not the whole SDM) | references/sdm-tool-reference.md |
"create a viz / chart" OR which-chart-type / create_visualization payload (no existing chart; not "just advise") | references/tasks/create-viz.md (then that guide cites viz-authoring.md — do not skip to it) |
| "edit / add a filter / forecast / sort / field to this chart" (an existing viz) | references/tasks/edit-viz.md |
| "show / view / display my existing chart/visualization" (no edit) | Call get_visualization directly (minorVersion: -1) — renders inline; do NOT chain render_visualization |
| "show / view / display / open my existing dashboard" (no edit) | Call render_dashboard (addVisualization=true, addSdm=true, showDraft=true, minorVersion=-1) — details in references/dashboard-authoring.md §8. Do not use get_dashboard (metadata-only, does not render) |
| "create a dashboard" / design a dashboard layout / add a global filter (no existing dashboard to edit) | references/tasks/build-dashboard.md |
| "edit an existing dashboard" / add-move-remove a widget / add-rename-delete a page on a dashboard that already exists | references/tasks/edit-dashboard.md |
| "how should I design the narrative/layout before charting" (no existing dashboard to review) | references/dashboard-design-principles.md |
| share / revoke / list who has access (user or group on a dashboard, viz, workspace, or SDM) | references/tasks/share-asset.md |
| reuse a prod asset in a Personal Org or promote a Personal Org asset / open a promotion PR | references/tasks/promote-or-reuse.md |
| provision / list Tableau Next users / check licenses (org role, not asset ACL) | references/tasks/provision-user.md |
| "alert me when" / change that alert / list or delete my alerts (ongoing monitor) | references/tasks/manage-alert.md |
| list / classify verified questions or score agent SQL against expected SQL | references/tasks/review-verified-questions.md |
Destructive / lifecycle operations — component and whole-asset deletes,
workspace asset association, share/reuse/promotion mutations, upsert_user,
delete_alert, delete_analytics_utterances, edit_dashboard drafts, and
double-wrapped defaultExc responses: read
references/destructive-operations.md before calling any of them. Always
get explicit user confirmation before delete_visualization /
delete_dashboard / delete_semantic_model / delete_workspace, and never
call save_dashboard or upsert_user without the user's explicit go-ahead.
Disambiguation:
- calculated vs. raw — a calc dimension/measure/metric routes to
enrich-model.md; a plain object/dimension/measure/join/logical view routes toedit-sdm.md. - analyze vs query vs greenfield — figures (counts, totals, scalars;
"how many orders", "what is total revenue") →
query-model.md(run_semantic_query; discover the SDM withlist_semantic_modelsif unnamed), including when the user says not to build or change a model. Interpreted breakdowns, rankings, trends, comparisons →analyze-data.md(analyze_data; do not pintargetEntityNameOrId/targetEntityTypeunless the user named an SDM). Named already-built model + structured table / explicitrun_semantic_query→query-model.md. Raw SQL against a DLO/DMO (run_query) when no SDM covers the data, or the user asked for SQL, also →query-model.md(that guide prefersrun_semantic_querywhen an SDM exists). Nothing modeled yet / CSV / "build a dashboard" →build-end-to-end.md. Ingest from an existing database connection (not a local file) still →ingest-flat-file.md. Derived table / Data Cloud SQL transform →create-data-transform.md, not ingest and notenrich-model.md. Metadata discovery (list SDMs, fields, relationships) islist_*/get_*, notanalyze_data. - create vs. edit a viz — a request that names or clearly references an
existing chart ("add a forecast to my monthly sales chart", "filter this
chart to West") routes to
edit-viz.md; a request with no existing chart in play ("create a bar chart of sales by category", "which chart type and whatcreate_visualizationpayload for a stacked bar") routes tocreate-viz.md. Do not skip toviz-authoring.md. - create vs. edit a dashboard — a request that names or clearly references an
existing dashboard ("add a KPI widget to my sales dashboard", "move this
chart to the top", "add a details page to the pipeline dashboard") routes to
edit-dashboard.md; a request with no existing dashboard in play ("create a dashboard for regional sales", "design a dashboard layout for these metrics") routes tobuild-dashboard.md. Do not skip todashboard-authoring.md. - show vs get vs edit a dashboard — "show / open / view my Sales
dashboard" (including naming or id'ing one to see it), no edit →
render_dashboard(dashboard-authoring.md§8). Widget JSON / PATCH oredit_dashboardprep on an existing dashboard →get_dashboardthenedit-dashboard.md. "Review this dashboard, don't modify it" stays No-dispatch. - Fast-path — if a suitable already-built, populated, joined model is
available (named, or discovered via
list_semantic_models) and the user wants figures or a structured table, this skill's build steps are not the entry point: route toquery-model.md(run_semantic_query). An interpretive question (breakdown, ranking, trend, comparison) →analyze-data.md. - Greenfield — "analyze account engagement / build a dashboard" where no
model exists yet is a build intent (the trigger is "the user wants insight
from data that isn't modeled yet," not the literal phrase "semantic model") →
build-end-to-end.md. - share vs workspace membership — granting a user/group
accessTypeon an asset routes toshare-asset.md. Putting an asset in a workspace (add_workspace_asset) is not a share. - revoke share vs delete the asset — "unshare Alice" / "remove her access"
→
share-asset.md. Deleting the dashboard/model/workspace itself stays on the destructive tools above. - reuse / promote vs create — copying or packaging an existing asset
routes to
promote-or-reuse.md. Building a new SDM from data still goes tobuild-end-to-end.md/edit-sdm.md. - share vs provision a user — "share this dashboard with Alice" /
viewer-vs-editor on an asset →
share-asset.md(search_users_and_groups). "Add Alice as Analyst" / list Tableau Next users / check seats →provision-user.md(get_users/upsert_user). Granting a role is not grantingaccessTypeon an asset. - analyze vs alert — one-off "why did EMEA sales drop" /
breakdown/trend/comparison →
analyze-data.md. "Alert me when bookings drop 10% week over week" / change or list my alerts →manage-alert.md. Do not pin an SDM on either path. - query vs score SQL — "what is total revenue" →
query-model.md. "Does this agent SQL match the expected SQL" →review-verified-questions.md(run_regression_evaluator; persists nothing). - AI-ready vs verified questions — backfill descriptions /
agentEnabled/ prefs →ai-readiness-audit.md. List/classify AVQ or score golden Q&A →review-verified-questions.md. - insights (analyze vs metric-card payload) — "generate insights on EMEA
sales" / interpretive "what should I focus on" →
analyze-data.md. The dashboard metric-card live value/trend/status payload isgenerate_insight_bundle/render_metricfor the MCP app renderer (insight-bundle.md) viaapp.callServerTool(). The model does not call those tools on a user chat turn and does not first-hop toinsight-bundle.md.
Show full SKILL.md (783 more words)Show less
No-dispatch rule (decline — do NOT route into an authoring guide)
The surface verbs above ("create a viz", "edit sdm", "create metric", "create/update dashboard", "share", "promote", "reuse", "provision", "alert", "classify") collide word-for-word with work this skill is not for. Classify on build-vs-advise/review/ideate/edit-prose, not the surface verb. Decline (or hand off) these — do not dispatch:
- Advice, no build — "which chart type, stacked vs grouped? don't create
anything" → advise; do not route to
create-viz.md. - Prose edit on an existing model — decline any request whose only change
is the wording of an existing description/label, no matter the verb used
("rewrite", "update", "fix", "correct", "improve", "reword" all count — the verb
is not the test). The test is: does this add/change a field, join, metric, or
grain? If not — pure copy-editing — decline. E.g. "rewrite the description on
my existing model" or "fix the typo in this field's description" → decline. This
is the sharpest near-miss:
edit-sdm.md/ai-readiness-audit.mdliterally wrapupdate_semantic_model_dimension/_measureas sparse description updates, so it is tempting — but a prose-only edit is not authoring the model. (Backfilling missing readiness metadata as part of making a model AI-ready IS in scope and must echoisVisible+dataObjectFieldName— see G5; rewording existing prose for style, accuracy, or typos is not.) - Ideation, no data — "brainstorm KPIs I could track later, no changes now" →
ideate; do not route to
enrich-model.md. - Review — "review my dashboard layout, don't modify it" / "someone on my
team built this, what's confusing" → review; do not route to
build-dashboard.md. Designing the narrative before you drop charts is the opposite intent — that does dispatch (todashboard-design-principles.md). - Share advice, no grant — "who should I share this with / viewer vs
editor? don't actually share" → advise; do not route to
share-asset.md. - Provision advice, no mutate — "who should I add as Analyst? don't
provision anyone" → advise; do not route to
provision-user.md. - Alert advice, no monitor — "should I alert on this KPI? don't create
one" → advise; do not route to
manage-alert.md.
A "don't build / don't change anything" disclaimer does not by itself
mean No-dispatch. Count / total / scalar ("how many orders") and named-model
structured query still route to query-model.md; interpreted breakdown /
ranking / trend / comparison still route to analyze-data.md. Those are
in-scope redirects, not declines.
Also out of scope: debugging a broken MCP server (→ Columbo / standard debugging); writing application code / managing AI Suite infra.
Deep references (cited by task guides)
The task guides point down to these for verified payloads and tool shapes.
A narrow lookup — "how do I / what is / what's the exact ___" — is a
direct-dispatch only when the list below names that file. Viz,
dashboard-widget, ingest, and greenfield-build lookups are not
direct-dispatch: first-hop to the matching references/tasks/*.md guide
(the guide then cites the deep ref). In particular, chart-type /
create_visualization payload questions first-hop to tasks/create-viz.md;
viz-authoring.md is not a first hop.
Direct-dispatch lookups:
- "how do I handle a huge/wide CSV with cryptic columns" →
large-flat-file-handling.md - "how do I figure out grain / measures vs. dimensions [before modeling]" →
data-understanding.md - "what's the input shape / suffix rule for
add_semantic_model_*" →sdm-tool-reference.md - "how do I bind SDM fields in one
create_semantic_modelcall" →sdm-tool-reference.md - "how do I add or update a semantic-model parameter" →
tasks/edit-sdm.md(shapes insdm-tool-reference.mdParameters) - "how do I add a HardJoin / Union / CustomSQL logical view" →
tasks/edit-sdm.md(shapes insdm-tool-reference.md) - "how do I safely delete a calc / metric / data object / logical view / relationship" →
sdm-tool-reference.md - "what depends on this viz / dashboard" / before deleting a viz or dashboard →
dashboard-authoring.md - "how do I show / open / view a dashboard" (inline render, not metadata) →
dashboard-authoring.md(§8render_dashboard) - "why did my query return zero rows" / "empty vs. does-not-exist" →
empty-source-handling.md - "how do I write a
run_semantic_querybody" →semantic-query-and-enrichment.md - "how do I make my model AI-ready" (checklist; not "backfill it now") →
ai-readiness.md - "how should I design the narrative/layout before charting" →
dashboard-design-principles.md
Not a first hop (cited only from the matching task guide):
- "which chart type / what
create_visualizationpayload" →tasks/create-viz.md(NOTviz-authoring.md) - "how do I add/move/remove a widget on an existing dashboard" →
tasks/edit-dashboard.md(NOTedit-dashboard.mdthe deep ref, and NOTdashboard-authoring.md) - "how do I create a dashboard / lay out a new dashboard" →
tasks/build-dashboard.md(NOTdashboard-authoring.md) - "how do I ingest a CSV / Excel / existing connection into a DLO/DMO" →
tasks/ingest-flat-file.md(NOTingest-and-metric-gotchas.md) - "how do I create or update a data transform" →
tasks/create-data-transform.md(NOTdata-transform-gotchas.md) - "walk me through building a model end to end" →
tasks/build-end-to-end.md(NOTbuild-workflow.md) - "how do I share / unshare / list who has access" →
tasks/share-asset.md(NOTsharing-and-promotion.md) - "how do I reuse a prod asset or promote a Personal Org asset" →
tasks/promote-or-reuse.md(NOTsharing-and-promotion.md) - "how do I provision / list Tableau Next users / check licenses" →
tasks/provision-user.md(NOTadmin-users.md) - "how do I create / update / list / delete a data alert" →
tasks/manage-alert.md(NOTalerts.md) - "how do I list verified questions / score agent SQL" →
tasks/review-verified-questions.md(NOTverified-questions.md)
What each deep reference covers (and which guide cites it) is in
references/deep-reference-index.md.
© forcedotcom, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 64 other files (references) in skills/tableau-next-asset-create of forcedotcom/sf-skills.
- SKILL.md
- README.md
- references/admin-users.md
- references/ai-readiness.md
- references/alerts.md
- references/build-workflow.md
- references/dashboard-authoring.md
- references/dashboard-design-principles.md
- references/data-transform-gotchas.md
- references/data-understanding.md
- references/deep-reference-index.md
- references/destructive-operations.md
- references/edit-dashboard.md
- references/edit-visualization.md
- references/empty-source-handling.md
- references/ingest-and-metric-gotchas.md
- references/insight-bundle.md
- references/large-flat-file-handling.md
- references/schemas/edit-dashboard
- … and 46 more
Open the folder on GitHubat commit e5164d9
Compare with similar skills
Tableau Next Asset Create next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Tableau Next Asset Create this skillforcedotcom/sf-skills | 1.1k | — | ~6.1k | Automated safety check: Pass | Apache-2.0 | |
| Mindsdb MCP SkillLeoYeAI/openclaw-master-skills | 2.2k | — | ~2.5k | Automated safety check: Pass | MIT | |
| Excel and CSV Data Analysisbytedance/deer-flow | 84k | 4 repos | ~2.2k | Automated safety check: Pass | MIT | |
| Rust SQL Testshencangsheng/easydb_app | 590 | — | ~1.2k | Automated safety check: Pass | MIT | |
| Data AnalysisHezaoHezao/poirot | 249 | — | ~997 | Automated safety check: Pass | MIT | |
| DatalionHybridAIOne/hybridclaw | 158 | — | ~2.8k | Automated safety check: Pass | MIT |
Similar skills
Excel and CSV Data Analysis
bytedance/deer-flow
Analyzes uploaded Excel and CSV files with SQL through DuckDB, producing schema inspections, statistical summaries and exports to CSV, JSON or Markdown.
Rust SQL Test
shencangsheng/easydb_app
Enforces unit test requirements for the Rust data-processing modules in src-tauri/src/sql/ (generator.rs, parse.rs) and src-tauri/src/reader/ (excel.rs and other readers).
Data Analysis
HezaoHezao/poirot
Analyze Excel/CSV files with DuckDB SQL via bash. An agent skill from HezaoHezao/poirot.
Datalion
HybridAIOne/hybridclaw
A skill your agent uses for DataLion workflows such as listing, reading, creating, or editing projects, inspecting data sources, importing Excel or CSV data, working with reports and report tabs and…
Tableau
LeoYeAI/openclaw-master-skills
Expert on modern configuration converter — converts Excel/CSV/XML/YAML into Protobuf-backed JSON/Text/Bin configs via protogen + confgen pipeline.
More from forcedotcom/sf-skills
All 251 skills in this repoAgentforce Architecture Analyze
forcedotcom/sf-skills
Declared architecture snapshot for one Agentforce agent: planner, topics, actions, flows, Apex, prompt templates, and NGA plugins.
Agentforce D360 Analyze
forcedotcom/sf-skills
Data Cloud 360° view of a single Agentforce session. An agent skill from forcedotcom/sf-skills.
Automation Sandbox Post Copy Configure
forcedotcom/sf-skills
Apply a Salesforce sandbox post-copy automation JSON config against a target org.
Automation Sandbox Post Copy Configure
forcedotcom/sf-skills
Apply a Salesforce sandbox post-copy automation JSON config against a target org.
Design Systems Slds Apply
forcedotcom/sf-skills
Apply SLDS-compliant UI using the correct blueprints, styling hooks, utility classes, and icons.
Experience Lwc Generate
forcedotcom/sf-skills
Lightning Web Components with PICKLES methodology and 165-point scoring.
Related
Categories
Questions about Tableau Next Asset Create
What does Tableau Next Asset Create do?
Build and edit Tableau Next semantic models (SDMs), vizzes, and dashboards via MCP. Tableau Next Asset Create is an agent skill from forcedotcom/sf-skills. Build and edit Tableau Next semantic models (SDMs), vizzes, and dashboards via MCP.
When should I use Tableau Next Asset Create?
Tableau Next Asset Create fits situations like: : building a semantic model; dashboard from CSV/Excel files; ingesting a flat file/database into DLO/DMO; profiling a dataset (grain.
How do I install Tableau Next Asset Create in Claude Code?
Run `npx skills add forcedotcom/sf-skills --skill tableau-next-asset-create -a claude-code`. Or copy the skill folder (skills/tableau-next-asset-create in forcedotcom/sf-skills) into .claude/skills/tableau-next-asset-create in your project. Claude Code loads it when a task matches its description.
How do I install Tableau Next Asset Create in Codex?
Run `npx skills add forcedotcom/sf-skills --skill tableau-next-asset-create -a codex`. Or copy the skill folder (skills/tableau-next-asset-create in forcedotcom/sf-skills) into .agents/skills/tableau-next-asset-create in your project. Codex loads it when a task matches its description.
Can I use Tableau Next Asset Create in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add forcedotcom/sf-skills --skill tableau-next-asset-create -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/tableau-next-asset-create, .gemini/skills/tableau-next-asset-create, .github/skills/tableau-next-asset-create and .opencode/skills/tableau-next-asset-create in your project.
What does Tableau Next Asset Create need to run?
SKILL.md names no scripts, command-line tools or credentials: Tableau Next Asset Create is instructions for the agent only.
Does Tableau Next Asset Create access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Tableau Next Asset Create safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Tableau Next Asset Create use?
Tableau Next Asset Create is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Tableau Next Asset Create use?
About 6.1k tokens (SKILL.md is roughly 24k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 134k tokens, read only when the agent opens those files.
What are the alternatives to Tableau Next Asset Create?
Skills that share tags, products or a category with Tableau Next Asset Create: Mindsdb MCP Skill (LeoYeAI/openclaw-master-skills, 2.2k stars), Excel and CSV Data Analysis (bytedance/deer-flow, 84k stars), Rust SQL Test (shencangsheng/easydb_app, 590 stars) and Data Analysis (HezaoHezao/poirot, 249 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Tableau Next Asset Create?
forcedotcom (a GitHub organization) maintains it in forcedotcom/sf-skills, which has 1,065 GitHub stars. The repository holds 251 skills in this directory. The repository was last updated on October 7, 2026.
Source: forcedotcom/sf-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.