search_notesSearches notes you can read. Filters: query (substring, case-insensitive), tag (OBSERVATION/PROBLEM/INSIGHT/BUG), utId, sessionId, useCaseId. Returns up to 200 matches with use-case context so you can group findings per use case.
| Name | Required | Description | Default |
|---|---|---|---|
| tag | No | ||
| utId | No | ||
| query | No | ||
| sessionId | No | ||
| useCaseId | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations only provide readOnlyHint=true. The description adds valuable behavioral details: substring matching is case-insensitive, results cap at 200, and returned data includes use-case context. These go well beyond the annotation and inform agent expectations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two sentences, front-loaded with the core purpose. Filters are neatly formatted in backticks, and the return behavior is stated compactly. No wasted words.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a search tool with 5 optional parameters and no output schema, the description covers purpose, filter semantics, result limit, and return context. It lacks exact definitions of the ID parameters and precise return structure, but remains largely complete for an AI agent.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description is the sole source for parameter meaning. It explains `query` (substring, case-insensitive) and `tag` (enum values), but `utId`, `sessionId`, and `useCaseId` are only named with no explanation of what they represent or how they filter. This leaves significant gaps.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('Searches') and resource ('notes you can read'), adding a scope limitation. It distinguishes itself from sibling getters like get_ut or get_session by being a cross-note search, not a single-entity retrieval.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Provides clear context: supports filtering by query/tag/IDs and returns use-case context for grouping. However, it does not explicitly state when to avoid this tool or mention alternatives like list_uts or get_session as substitutes for specific lookups.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Every tool targets a distinct resource and action. Even get_ut, which returns comprehensive data, doesn't replace the specialized get_session, get_transcript, or get_quantitative_report for detailed views. No two tools have overlapping purposes.
All names follow a verb_noun pattern with underscores (e.g., add_task, get_ut, list_workspaces). The main inconsistency is using 'create' for UTs (create_ut) but 'add' for child entities (add_use_case, add_task), and having a separate update_scenario instead of folding it into update_ut, but the convention is still predictable.
With 18 tools, the set is slightly above the typical 3-15 well-scoped range, but the domain of usability testing requires distinct operations for drafting (use cases, tasks, scenario), reading results (reports, sessions, transcripts), and comparison. Each tool has a clear role, so the count feels reasonable rather than excessive.
The tool surface covers the core lifecycle: create, read, update, and delete for tasks/use cases, plus comprehensive retrieval of results and notes. Minor gaps include no delete_ut (UTs cannot be removed) and no explicit status change or cloning mechanism, but these are likely intentional app-side actions, so the surface is not severely incomplete.