UNDERSTANDING — Delete a saved search you own.
| Name | Required | Description | Default |
|---|---|---|---|
| id | Yes | ||
| context | No | Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are present, so the description carries the full burden of behavioral disclosure. It reveals that the operation is destructive and requires ownership, but it omits critical details like permanence, failure behavior for nonexistent or unauthorized searches, and any cascading side effects.
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 a single, direct sentence with no filler. The 'UNDERSTANDING —' prefix is extraneous and adds no semantic value, which prevents a perfect score for conciseness.
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?
This is a destructive tool with no annotations and no output schema, yet the description provides no information about success/failure behavior, reversibility, or error reporting. The agent has insufficient context to invoke it safely and predict outcomes.
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 50%, with only the context parameter described. The tool description adds meaning by clarifying that 'id' refers to the saved search being deleted, but it does not elaborate on id format, error handling, or the exact role of the id beyond the obvious.
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 states a specific verb ('Delete') on a concrete resource ('a saved search') with an ownership constraint ('you own'). This clearly differentiates it from sibling tools like delete_list and other saved-search helpers, making the tool's intent unmistakable.
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?
The description implies when to use the tool: when deleting a saved search the user owns. However, it does not explicitly mention alternatives (e.g., delete_list) or exclusion scenarios, leaving the usage guidance mostly implicit rather than explicit.
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.
Most tools are clearly separated by artifact type or resource (find_mcp vs find_openapi vs get_provider vs get_api), but the sheer volume creates some genuinely confusable clusters: apis_io_search vs find_apis vs find_artifacts, and insights_adoption vs insights_dimensions vs find_company_insights. Several readiness-related tools (what_can_i_fix, simulate_fixes, readiness_gates) also share a conceptual boundary, though their descriptions do help.
The dominant patterns (find_*, get_*, cohort_*, compare_*) are consistent and predictable, but the set mixes in irregular names like apis_io_search, tag_group_tags, what_can_i_fix, whats_changed, and resolve. These deviations are readable but break the otherwise regular verb_noun convention.
106 tools is far beyond the typical well-scoped server and will impose a heavy selection burden on agents. The server covers a genuinely broad domain (catalog search, ratings, cohorts, agent readiness, lists, exports, feedback), so the count is defensible in scope, but it is still too many to navigate efficiently.
The surface is remarkably complete: search and browse, single-entity detail, comparisons, cohort analytics, agent-readiness assessment, saved searches, list management, feedback/correction flows, and full dataset exports are all covered. There are no obvious dead ends, and even minor operations like re-running saved searches or simulating fixes are present.
