collaboration-platform-advisor
Collaboration platform configuration methodology for legal matter sites. Site architecture, workflow identification, dashboard design, data quality governance, and user adoption for SharePoint, Teams, and equivalent platforms. M365 is the reference implementation — outputs are platform-agnostic enough to brief IT or build simple automations without becoming a Power Automate manual. Use when setting up a matter site, identifying workflows to automate, designing reporting dashboards, managing platform data quality, or driving user adoption. Trigger on: 'set up the matter site', 'configure SharePoint', 'build a dashboard', 'what should we automate', 'brief IT on this workflow', 'nobody is using the platform', 'data quality is poor', 'set up Teams channel', 'matter site structure', 'alerts and notifications', 'user training', 'platform governance', 'status dashboard', 'what workflows can we automate', 'matter site template'.
68
81%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that clearly defines a specific niche (legal matter site configuration on collaboration platforms), lists concrete capabilities, and provides comprehensive trigger guidance. The explicit 'Use when' clause and extensive 'Trigger on' list with natural user phrases make it highly effective for skill selection. The scope boundary clarification ('without becoming a Power Automate manual') further aids distinctiveness.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: site architecture, workflow identification, dashboard design, data quality governance, user adoption, and specifies platforms (SharePoint, Teams, M365). Also clarifies scope boundaries ('without becoming a Power Automate manual'). | 3 / 3 |
Completeness | Clearly answers both 'what' (site architecture, workflow identification, dashboard design, data quality governance, user adoption for legal matter sites) and 'when' with an explicit 'Use when...' clause and a detailed 'Trigger on:' list covering multiple scenarios. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms users would actually say, including 'set up the matter site', 'configure SharePoint', 'build a dashboard', 'nobody is using the platform', 'data quality is poor', 'what should we automate', and many more realistic variations. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche: legal matter site configuration on collaboration platforms. The combination of legal domain specificity, platform focus (SharePoint/Teams/M365), and concrete trigger terms like 'matter site' and 'matter site template' make it very unlikely to conflict with generic SharePoint or dashboard skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
62%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a highly actionable and well-structured skill with excellent concrete templates, clear validation gates, and strong workflow sequencing across all four modes. Its primary weakness is significant verbosity — the skill is roughly 3-4x longer than it needs to be, with extensive design philosophy, repeated behavioral instructions, and domain knowledge sections that restate principles already embedded in the mode instructions. The monolithic structure compounds the token cost problem.
Suggestions
Extract the 'Domain Knowledge — Why Law Firm Platforms Fail' section into a separate reference file — these principles are already operationalized in the mode instructions and the section is redundant.
Consolidate repeated output rules ('produce the output — do not ask whether to produce it', 'use placeholders', 'flag gaps at the end') into a single shared instruction block rather than restating variants in each mode and in the Output Format section.
Move the detailed automation candidate table, diagnostic tables, and dashboard specification example into bundle reference files, keeping only a brief summary and pointer in the main SKILL.md.
Remove explanatory prose that restates what the templates already encode — e.g., the paragraph explaining why status report compilation requires judgment can be reduced to a single line in the 'What not to automate' list.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~400+ lines. It explains design philosophy extensively ('Law firm platform failures are rarely technical — they are design and adoption failures'), includes a full 'Domain Knowledge — Why Law Firm Platforms Fail' section that restates principles already embedded in the modes, and repeats output rules and behavioral instructions multiple times. Much of the methodology explanation (e.g., why platforms fail, what not to automate) is knowledge Claude can infer from context. The M365 Connected Mode section and Time-Sensitive Assumptions add further bulk with limited actionable value. | 1 / 3 |
Actionability | The skill provides highly concrete, copy-paste-ready output templates (automation brief skeleton, dashboard specification, adoption intervention plan, identifier block, pre-flight checklist), specific SharePoint list schemas with named fields, a detailed automation candidate table with triggers/actions/values, and structured diagnostic tables for data quality and adoption issues. Every mode has a clear output format that can be populated immediately. | 3 / 3 |
Workflow Clarity | Each mode has a clear sequence: hard gate (identifier block) → pre-flight checklist → mode-specific inputs → structured output with explicit output rules. The hard gate is a genuine validation checkpoint ('do not produce any site architecture... until the identifier block is confirmed'). Mode 4 has a diagnosis-before-fix sequence. Mode 2 has a four-part test applied before any automation is approved. The output rules explicitly prevent common failure modes (asking instead of producing, withholding output pending information). | 3 / 3 |
Progressive Disclosure | The content is entirely monolithic — everything lives in a single long SKILL.md with no bundle files or external references. The automation brief templates, dashboard specifications, diagnostic tables, and domain knowledge sections could all be split into referenced files. The cross-skill handoffs section references other skills but the skill's own content is not decomposed. For a skill this long, the lack of any content splitting is a notable organizational weakness. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- legalopsconsulting/lpm-skills
- Commit
dc6509d
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.