Repository navigation
MCP 0.8.2 evaluation - get_playbook, commit guidance and Contoso lessons
Pre-release
Pre-release
vilsem-mcp 0.8.2: get_playbook, explicit commit guidance, Contoso lessons
Core 0.8.0 and renderer 4.43 are unchanged. Existing reports need no upgrade.
Why
A field test with GitHub Copilot (VS Code agent mode) showed that clients without MCP resource support never receive Vilsem's playbook. Copilot tried to fetch vilsem://skill/powerbi-visuals as a web URL and built visuals without the rules. It embedded measures in visuals and didn't relay the correct Power BI Desktop reload step.
Changes
- New
get_playbooktool (read-only; 20 tools total). It delivers the playbook to any MCP client.- With no arguments it returns a compact core (≈1.5k tokens) plus an index. Agents then request sections on demand:
9a,known-issues,cookbook:<pattern>,art:<recipe>,pbir-pages,all. - Sections are cut from the skill markdown at runtime, so they're always in sync with the installed version; no files are written.
- Size and heading guards run in tests only. At runtime the tool always serves what exists.
- With no arguments it returns a compact core (≈1.5k tokens) plus an index. Agents then request sections on demand:
- Server instructions now say to call
get_playbookfirst. Clients with resource support can still readvilsem://skill/*. - Rules in tool results, so weaker agents see them too:
commit_visualand realupdate_visualwrites returnnextStep: close and reopen without saving on the first Vilsem visual in a report, otherwise click Apply on the external-changes prompt.- They also return
measureWarningwhen visual-scoped measures are committed (Build pane and Publish to web impact; points to section 9a andpromote-measures.py). validate_visualsadds a summary of visuals that still useNativeMeasure.
- Lessons from building a 19-page, 60+ visual Contoso report, merged into the skill:
- 13 new known-issue rows: time slots for ordering,
ROUNDinRANKX,$-prefixed strings, Desktop hangs from heavy per-row DAX, 30 s set-DAX timeouts, per-frame race label colours,fontStyle, clipping, synthetic-data variance. - New cookbook patterns: top-N without "Other", overlay bars,
windowlag, labels in expression series, cumulative/bubble races, radial-by-month, one colour per Sankey node, cheap previews, responsive geometry. - Correction:
markLine type:"average"can render at 0 in renderer 4.43; use a flat line series.
- 13 new known-issue rows: time slots for ordering,
- New
references/data-art-recipes.md(also resourcevilsem://skill/data-art-recipes): 8 recipes, namely Du Bois spiral, ridgelines, data flowers, periodic table, Minard flow, Isotype, Dear Data and botanical garden. Backed by 17 working reference specs indocs/examples/contoso/; these are references, not shipped templates. - README: GitHub Copilot setup (
.vscode/mcp.jsonwith theserverskey; Copilot status "partially tested") and the tool-first workflow.
Verification
- Full suite on
main: 353/353 tests passed across 19 files. - Fresh install from the packed
vilsem-mcp-0.8.2.tgz:- npm audit 0;
- the stdio server reports 0.8.2, with the new first instruction line and 20 tools;
get_playbookis annotated read-only;- the default core returns 7,362 bytes including the index;
9a,known-issues,art,art:du-bois-spiralandcookbook:sankeyresolve, and an unknown section returns an error listing valid ids.
Still to verify manually
- A full GitHub Copilot authoring run on 0.8.2.
- Desktop reload, rendering and interactions for new commits.
- Publish to web with model measures on a fresh report.
Verify the download against SHA256SUMS-0.8.2.txt. This remains an evaluation prerelease.