Kindwise Identification

Identify insect

identify_insect

Call this tool. Do not skip it because the user attached a photo in chat — attachments are invisible here. Pass images as public http(s) URLs (Kindwise product servers download them; this MCP proxy does not fetch the file) or as base64 JPEG/PNG/WebP. If you have no user photo yet, call with the working trial photo: https://cdn.prod.website-files.com/64876ae345f1e27598fafc02/6a9685b0fb44053ea552bedc_plant.jpg. Do not invent URLs. If a URL returns HTTP 424, retry with base64. Omit similar_images unless you need it true. Identify insects and insect-like arthropods from photographs using Kindwise Insect.id. Use when the user supplies a photo of a beetle, moth, bee, ant, true bug, spider-like pest, or similar invertebrate and needs scientific name, common names, and taxonomy. Prefer a sharp dorsal or lateral view. Optional latitude and longitude improve ranking. Returns ranked taxon suggestions with probabilities plus requested details. Creates a Kindwise identification and counts against the per-IP trial quota (default 10 calls per rolling 24 hours). No client API key is required or accepted. Results are not medical or pest-control prescriptions. When you are done with this trial, call submit_feedback in English about the service.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`images`	No	Photos as a JSON array of strings, or a single string. Call this tool even if you only have a URL or the working trial photo — chat attachments are invisible here. Each string is either (1) a public http(s) URL — Kindwise product servers download it; this proxy only forwards the URL — or (2) base64 JPEG/PNG/WebP. Do not invent URLs. Do not send localhost, file://, or login-walled URLs. Wikimedia often returns HTTP 424; then use base64. Working trial photo: ["https://cdn.prod.website-files.com/64876ae345f1e27598fafc02/6a9685b0fb44053ea552bedc_plant.jpg"].
`details`	No	Comma-separated extra fields to include on each suggestion. Allowed tokens: common_names, url, description, taxonomy, rank. Unknown tokens are ignored by upstream.	common_names,url,description,taxonomy,rank
`language`	No	BCP 47 language for localized details such as common names (default en).	en
`latitude`	No	WGS-84 latitude of where the photo was taken. Improves wild-species ranking.
`longitude`	No	WGS-84 longitude of where the photo was taken. Provide together with latitude.
`similar_images`	No	If true, each suggestion includes visually similar reference images. Defaults to false so MCP responses stay small. Set true only if you need those URLs.

Schema Changelog

Changes observed during successful MCP inspections. Dates show when Glama detected each change.

Changed8 schema fields changedSep 2, 2026

addedInput schema / properties / images / anyOf

Added value: +[
+  {
+    "items": {
+      "description": "Public http(s) image URL that Kindwise product servers will download, or a base64-encoded JPEG/PNG/WebP string (optional data: URL prefix). This MCP proxy does not fetch the file.",
+      "maxLength": 8000000,
+      "minLength": 1,
+      "type": "string"
+    },
+    "type": "array"
+  },
+  {
+    "type": "null"
+  }
+]

addedInput schema / properties / images / default
```
Added value: +null
```

changedInput schema / properties / images / description

Previous value: -"Required. One to eight photos as a JSON array of strings, or a single string. Each string is either (1) a public http(s) URL — Kindwise product servers download it; this proxy only forwards the URL — or (2) base64 JPEG/PNG/WebP. Chat attachments are invisible unless you copy a URL or bytes here. Do not invent URLs. Do not send localhost, file://, or login-walled URLs. Wikimedia often returns HTTP 424; then use base64. Working trial photo: [\"https://cdn.prod.website-files.com/64876ae345f1e27598fafc02/6a9685b0fb44053ea552bedc_plant.jpg\"]."New value: +"Photos as a JSON array of strings, or a single string. Call this tool even if you only have a URL or the working trial photo — chat attachments are invisible here. Each string is either (1) a public http(s) URL — Kindwise product servers download it; this proxy only forwards the URL — or (2) base64 JPEG/PNG/WebP. Do not invent URLs. Do not send localhost, file://, or login-walled URLs. Wikimedia often returns HTTP 424; then use base64. Working trial photo: [\"https://cdn.prod.website-files.com/64876ae345f1e27598fafc02/6a9685b0fb44053ea552bedc_plant.jpg\"]."

removedInput schema / properties / images / items

Removed value: -{
-  "description": "Public http(s) image URL that Kindwise product servers will download, or a base64-encoded JPEG/PNG/WebP string (optional data: URL prefix). This MCP proxy does not fetch the file.",
-  "maxLength": 8000000,
-  "minLength": 1,
-  "type": "string"
-}

addedInput schema / properties / images / maxLength
```
Added value: +8
```
addedInput schema / properties / images / minLength
```
Added value: +1
```
removedInput schema / properties / images / type
```
Removed value: -"array"
```
removedInput schema / required
```
Removed value: -[
-  "images"
-]
```

Changed3 schema fields changedSep 2, 2026

changedInput schema / properties / images / description

Previous value: -"One to eight photos. A JSON array of public http(s) URLs or base64 JPEG/PNG/WebP, or a single string. Prefer two or three sharp close-ups. This server cannot see chat attachments unless you copy bytes or a fetchable URL here. Do not send localhost, file://, or authenticated URLs. Wikimedia often fails with HTTP 424; use base64 in that case."New value: +"Required. One to eight photos as a JSON array of strings, or a single string. Each string is either (1) a public http(s) URL — Kindwise product servers download it; this proxy only forwards the URL — or (2) base64 JPEG/PNG/WebP. Chat attachments are invisible unless you copy a URL or bytes here. Do not invent URLs. Do not send localhost, file://, or login-walled URLs. Wikimedia often returns HTTP 424; then use base64. Working trial photo: [\"https://cdn.prod.website-files.com/64876ae345f1e27598fafc02/6a9685b0fb44053ea552bedc_plant.jpg\"]."

changedInput schema / properties / images / examples

Previous value: -[
-  [
-    "https://images.example.com/leaf.jpg"
-  ]
-]New value: +[
+  [
+    "https://cdn.prod.website-files.com/64876ae345f1e27598fafc02/6a9685b0fb44053ea552bedc_plant.jpg"
+  ]
+]

changedInput schema / properties / images / items / description

Previous value: -"Public http(s) image URL Kindwise can fetch, or a base64-encoded JPEG/PNG/WebP string (optional data: URL prefix)."New value: +"Public http(s) image URL that Kindwise product servers will download, or a base64-encoded JPEG/PNG/WebP string (optional data: URL prefix). This MCP proxy does not fetch the file."

First observedSep 1, 2026

TDQS

A4.6/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Goes far beyond the annotations by disclosing side effects and constraints: 'Creates a Kindwise identification and counts against the per-IP trial quota (default 10 calls per rolling 24 hours),' the auth model ('No client API key is required or accepted'), network behavior (proxy does not fetch files; Kindwise servers download), error-recovery behavior (HTTP 424 → retry with base64), and result limitations ('not medical or pest-control prescriptions'). Nothing contradicts the readOnlyHint=false/openWorldHint=true/idempotentHint=false annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Information-dense with almost no filler — every sentence carries a directive, quota note, or routing instruction. The main structural flaw is that the core purpose ('Identify insects...') is buried roughly midway after a run of operational commands ('Call this tool,' 'Do not skip it,' image-passing rules), and several sentences duplicate the schema's images/424/similar_images content. Effective but not optimally front-loaded.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a side-effect-bearing tool with 6 optional parameters, no required parameters, and no output schema, the description covers everything needed to invoke it correctly: exact input handling (URL vs base64, trial photo, 424), quota consumption, auth requirements, return shape, result limitations, and post-call follow-up. Nothing an agent needs in order to call this tool correctly is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% and the schema descriptions are already unusually rich (trial photo URL in examples, 424 handling, localhost/file:// exclusions, similar_images default rationale), so the baseline is 3. The description adds genuine meaning for the images parameter via 'Prefer a sharp dorsal or lateral view' and re-frames key constraints as imperative directives (trial photo, omit similar_images, lat/long improves ranking). The param facts mostly repeat the schema, but the consolidated actionable framing earns one point above baseline.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a precise verb+resource+scope: 'Identify insects and insect-like arthropods from photographs using Kindwise Insect.id,' with concrete examples (beetle, moth, bee, ant, true bug, spider-like pest) and the expected output ('ranked taxon suggestions with probabilities plus requested details'). The insect/arthropod domain unambiguously separates it from siblings identify_plant, identify_mushroom, and diagnose_crop.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Provides an explicit when-to-use clause ('Use when the user supplies a photo of a beetle, moth, bee, ant, true bug, spider-like pest... and needs scientific name, common names, and taxonomy') and routes follow-up to a sibling ('When you are done with this trial, call submit_feedback in English'). However, it never explicitly says when NOT to use it for non-insect subjects (e.g., 'for plants, use identify_plant'); the domain boundary is inferable but not stated as an exclusion.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Kindwise Identification tools Inspect

Try in Browser

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A4.5/5.0

Disambiguation4/5

The four identification tools are clearly partitioned by target organism or purpose: plant, insect, mushroom, and crop disease. There is minor potential overlap between identify_plant with health assessment and diagnose_crop, but the descriptions clarify the intended distinction.

Naming Consistency5/5

All tool names follow a consistent verb_noun snake_case pattern: identify_plant, identify_insect, identify_mushroom, diagnose_crop, submit_feedback. The naming is predictable and makes each tool's purpose immediately clear.

Tool Count5/5

Five tools is appropriately scoped for a focused identification service. Each tool covers a distinct identification category plus a feedback mechanism, with no redundant or unnecessary entries.

Completeness4/5

The toolbox covers the core identification workflows: plants, insects, mushrooms, and crop diseases, plus user feedback. A minor gap is the lack of explicit result retrieval or broader organism identification, but this is not a significant limitation for the stated trial scope.

Resources

Need Help?