Impression-weighted SERP position, click projection on your own CTR curve, share of voice
Scored across 3 tools
Each tool has a clearly distinct purpose: one computes weighted average position, one projects clicks from a user-supplied CTR curve, and one computes share of voice. There is no overlap that would confuse an agent.
All names use snake_case and are descriptive, but they mix noun phrases (impression_weighted_position, share_of_voice) with a verb phrase (project_clicks_from_ctr_curve). No consistent verb_noun pattern, yet names remain readable and predictable enough.
3 tools is well within the ideal 3-15 range. Each tool addresses a distinct analytics need and earns its place without bloat or gaps.
The set covers core SERP analytics: position aggregation, click projection, and share-of-voice. A minor gap is the lack of a direct CTR-computation utility, but the tools are complementary and cover the main workflows without dead ends.
Compute the impression-weighted average SERP position for a set of query rows, alongside the plain arithmetic mean and the gap between them. The arithmetic mean lets a 3-impression query count as much as a 30,000-impression one; the weighted figure describes where a typical impression actually happened. A positive weighting_shift means your high-volume queries rank worse than the unweighted average suggests.
| Name | Required | Description | Default |
|---|---|---|---|
| rows | Yes | Query rows. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It explains the mathematical behavior (weighted vs unweighted mean), defines weighting_shift, and interprets its meaning, which goes beyond a simple restatement. However, it does not mention any side effects, edge cases, or output structure.
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 three concise sentences, each providing meaningful information: the computation, the rationale, and the interpretation of a key output. No word is wasted, and the structure flows logically.
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?
The tool has a single parameter with a rich schema and no output schema. The description explains the core outputs conceptually and their interpretation, but it does not specify the exact return structure or handle potential edge cases. This is largely complete for a straightforward computation tool.
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?
The schema already documents all fields within the rows parameter with 100% coverage. The description does not add per-parameter details beyond what the schema provides, but it does provide context on how the data is used (e.g., positions and impressions). This matches the baseline for high schema coverage.
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 tool computes impression-weighted average SERP position, arithmetic mean, and the gap between them. It is specific and unambiguous, but it does not explicitly mention how it differs from sibling tools like share_of_voice or project_clicks_from_ctr_curve, so it lacks explicit sibling differentiation.
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 the tool is used when analyzing query rows with position and impression data, but it does not provide explicit when-to-use guidance or mention any alternatives. The context is understandable but no exclusions or clear alternatives are stated.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Project how many clicks a set of query rows would earn at a target position instead of their current position, using a CTR curve THAT YOU SUPPLY. This server deliberately ships no built-in CTR table: every published average-CTR-by-position table is a third-party estimate over someone else's traffic, so applying one silently would disguise a guess as a measurement. Pass your own measured curve (Search Console clicks/impressions by position is the honest source). Positions between curve points are linearly interpolated; positions beyond the curve clamp to the last point rather than extrapolate, and are counted in clamped_row_count. If any row carries observed clicks, the output also reports how far the supplied curve is from your reality.
| Name | Required | Description | Default |
|---|---|---|---|
| rows | Yes | Query rows. | |
| ctr_curve | Yes | Required. Map of position to click-through rate as a 0-1 fraction, e.g. {"1":0.28,"2":0.15,"3":0.10,"10":0.02}. Supply your own measured values. | |
| target_position | No | Target position applied to every row that does not carry its own. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden and does so admirably. It discloses that there is no built-in CTR table, that positions between curve points are linearly interpolated, that positions beyond the curve clamp and are counted in clamped_row_count, and that calibration output reports deviation from observed clicks. This goes far beyond a basic summary.
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 front-loaded with a clear purpose, followed by a concise rationale, guidance, and key behavioral details. Every sentence earns its place; the tone is direct and avoids fluff. At roughly 150 words, it is appropriately sized for the tool's complexity.
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 tool with nested objects, no annotations, and no output schema, the description covers all critical context: purpose, why there is no built-in curve, how to source the curve, interpolation/clamping behavior, and diagnostic output. It even names clamped_row_count and calibration reporting. This is well-rounded and complete.
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 100%, so baseline is 3. The description adds meaning by emphasizing that the ctr_curve must be user-supplied and based on real measurements, and it clarifies behavior around interpolation/clamping that connects to the rows and curve parameters. It does not elaborate on target_position semantics beyond what the schema offers, but the overall context is enhanced.
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 tool's function: 'Project how many clicks a set of query rows would earn at a target position instead of their current position, using a CTR curve THAT YOU SUPPLY.' This is specific and the verb 'project' plus the resource 'clicks from CTR curve' distinguishes it from siblings like share_of_voice or impression_weighted_position.
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 provides strong contextual guidance: it explicitly instructs the user to supply their own measured curve, warns against built-in tables ('every published average-CTR-by-position table is a third-party estimate'), and suggests Search Console as an honest source. It does not explicitly exclude alternatives among sibling tools, so it stops short of a 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
impression_weighted_positionproject_clicks_from_ctr_curveshare_of_voiceAI visibility: is your brand cited by ChatGPT, Perplexity, Gemini? SoV, GEO score, AI traffic.
AI-search visibility (GEO) analytics: brand scores, competitor rankings, recommendations, evidence.
Track AI visibility (ChatGPT, Gemini, AI Overviews), research SEO keywords, write, publish, prove.
Full SEO audits, content analysis, and domain overviews - on-page, technical, and authority metrics
Add one secure layer between your agents and this server.
