Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add ADIKANT/datalens-dev-mcp --skill datalens-editorgit clone --depth 1 https://github.com/ADIKANT/datalens-dev-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/adikant/datalens-dev-mcp/datalens-editor)<a href="https://agentmods.dev/skills/adikant/datalens-dev-mcp/datalens-editor"><img src="https://agentmods.dev/badge/skills/adikant/datalens-dev-mcp/datalens-editor/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/adikant/datalens-dev-mcp/datalens-editor"><img src="https://agentmods.dev/badge/skills/adikant/datalens-dev-mcp/datalens-editor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00049 | $0.00947 |
| Opus 5.5 | $0.00020 | $0.00379 |
| Sonnet 5.5 | $0.00010 | $0.00189 |
| Haiku 4.5 | $0.00005 | $0.00095 |
Grade A, and why
datalens-editor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 22d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 30 lines — stays where its author put it; the contents beside it link to each section on GitHub.
DataLens Editor
Use the technology and source selection order first.
For a known Editor, start with that target's current identity, parent and subtype/renderer; no workbook-wide inventory is needed. Use inventory when selecting an unknown target or discovering a suitable existing example. If that evidence is insufficient, inspect the business workspace's root navigation and only relevant project AGENTS.md/CONTEXT.md files; confirm a candidate's live ID and type. A Wizard-only project does not establish that Editor is unavailable. For authorized creation, use a supported recipe or example in the requested workbook, remapping aliases and dependencies to authorized objects or read-only sources. Dependencies do not expand mutation scope.
Work from the exact dashboard project/subproject. For a single-tab edit, read that target in full, patch only the requested tab, and verify the changed behavior; do not recompile the entire chart or traverse its workbook. Read Editor authoring for the actual runtime variant (table_node, d3_node, advanced-chart_node, markdown_node or control_node). Preserve that technology, untouched tabs, Meta aliases and source bindings. Diagnose missing aliases, upstream errors and valid empty results separately.
Use dl_compile_recipe for new registered visual families or a requested redesign, adjusting bindings and permitted presentation values. Its compact draft_reference addresses the complete private draft: pass it to validation/create; for update combine its artifact_path with exact target identity and fresh revision. A small existing-tab edit uses the observed tabs, not recipe compilation. Reuse packaged renderers rather than reproducing their JavaScript. New KPI/period/table recipes inherit widget title/hint ownership; pass a concrete bindings.hint and carry the resulting presentation into placement. Standalone pending placement does not create an internal replacement hint. Business labels, units and source IDs belong in bindings/project overrides.
dl_editor_validate checks static tabs, aliases and constrained-runtime rules. Verify changed runtime/visual behavior in Browser separately. HTML generated inside an Editor/table cell follows that Editor contract; standalone Page CSP and host messaging do not apply.
For period-series recipes, comparison must apply to the series: comparison.enabled=false disables it; an explicit true retains an unavailable state when previous data are missing. legend.mode=hidden hides the legend. Scope optional tooltip.hide_null, tooltip.hide_zero_multi and legend.hide_empty_series to the requested presentation; they do not change Dataset rows or share denominators. Follow the project's accepted profile and visualization decisions; existing-tab changes still use a narrow patch.
Follow authorized scope and delivery and, for a visual/semantic change, the relevant visualization decision. Route Wizard to datalens-dataset-wizard, placement to datalens-dashboard, discovery to datalens-inspect, and standalone Page/backup/cleanup to datalens-maintenance.
For source changes, use the effective presentation to identify active consumers. Period KPI comparison/sparkline flags determine previous-total/trend queries; value-only needs only the current aggregate. Other explicit per-alias Dataset projections use dataset_source in the same recipe compiler. Inspect summary.source_plan, actual DataSetData/SQL and results at the same grain. Projection does not prove reduced inner SQL cost. Follow query needs and portable presentation; preserve custom source code and metadata-only edit paths.
For unknown effects follow repeat-effect recovery: only proven non-application or separate informed repeat authorization permits another attempt. Failed reconciliation, absent inventory and a new operation_id do not make a retry safe.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 22d ago Changed · +7 lines 1dcb1810fc7a
- 29d ago Changed · +2 lines c88a21ecfa87
- 1mo ago Changed · +2 lines c09057540d7e
- 1mo ago Changed cfb20e1104d3
- 1mo ago Changed 3b32ce61ba3d
- 1mo ago Changed · +2 lines b719983b2e14
- 1mo ago Changed · +2 lines 897305ff4984
- 1mo ago Changed · -4 lines · +21 tokens per session f131a3044f6e
- 1mo ago Changed · +2 lines 4a055a2527fd
- 1mo ago First seen · 17 lines · 28 tokens per session scan A 650cd26e6614
datalens-editor is a skill published in the GitHub repository ADIKANT/datalens-dev-mcp (7 stars, last pushed 11d ago), licensed Apache-2.0. It adds 49 tokens to every session and 947 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-08.
Other skills, from other repositories
quantum-vqe
A runnable guide to the variational quantum eigensolver (VQE), a quantum algorithm that estimates the lowest energy of a mathematical system by adjusting a parameterized circuit. It uses Pauli Hamiltonians, a specified circuit form, and SciPy optimization.
quantum-vqls
A runnable guide to the variational quantum linear solver (VQLS), a quantum method that estimates a solution to a system of linear equations by optimizing a parameterized circuit. It uses Qiskit and reports the numerical residual and why optimization stopped.
octave-compression
Workflow for transforming prose into semantic OCTAVE structures. Covers tier selection, transformation phases, loss accounting, and decision rules. REQUIRES octave-literacy.
octave-chatter
OCTAVE on the wire — read and emit OCTAVE in agent-to-agent messages without authoring files. Syntax kernel, operator table, and the OCTAVEWIRE primer v1.3 (provenance, hedges, verbatim IDs). For agents outside the workbench matrix.
@elad12390/web-research-assistant
Comprehensive MCP server for web research with 16 tools: SearXNG + Exa AI search, batch search and batch crawling, URL crawling, stealth scraping (Cloudflare bypass), package registry lookup (npm/PyPI/crates), GitHub stats, changelog fetching, tech comparison, error translation, API docs discovery, stock image search…
inbox
Show the Relaydeck inbox and act on it. Use when the person asks what is new in Relaydeck, or asks for new tasks, answers, or messages from the app or from other agents.