Koskamo

Retrieve cited passages

koskamo_retrieve

Read-onlyIdempotent

Search Koskamo datasets for candidate passages with source citations and document versions. Similarity scores rank matches and do not measure factual confidence.

Input Schema

Name	Required	Description
`query`	Yes	The question or search string
`top_k`	No
`dataset`	No	Optional dataset id or slug. Defaults to all public datasets.

Output Schema

Name	Required	Description
`query`	Yes
`degraded`	No	Present only when a backend was unavailable and passages came from a keyword fallback; coverage may be partial
`guidance`	Yes	What returned passages and scores do and do not establish
`passages`	Yes
`assessment`	Yes	Candidate availability only; does not assess whether the question is answered

Schema Changelog

Changes observed during successful MCP inspections.

Changed2 schema fields changedOct 31, 2026

changedOutput schema / properties / guidance / description

Previous value: -"Evidence checking and abstention instructions for the consuming agent"New value: +"What returned passages and scores do and do not establish"

changedOutput schema / properties / passages / items / properties / datasetVersion / description

Previous value: -"Collection version at retrieval time; use documentVersion for an exact document citation"New value: +"Collection version at retrieval time; documentVersion identifies the exact document"

Changed1 schema field changedOct 20, 2026

addedOutput schema / properties / degraded

Added value: +{
+  "description": "Present only when a backend was unavailable and passages came from a keyword fallback; coverage may be partial",
+  "enum": [
+    "embeddings_unavailable",
+    "database_unavailable"
+  ],
+  "type": "string"
+}

Changed4 schema fields changedOct 8, 2026

addedOutput schema / properties / assessment

Added value: +{
+  "description": "Candidate availability only; does not assess whether the question is answered",
+  "enum": [
+    "candidates",
+    "no_matches"
+  ],
+  "type": "string"
+}

addedOutput schema / properties / guidance

Added value: +{
+  "description": "Evidence checking and abstention instructions for the consuming agent",
+  "type": "string"
+}

changedOutput schema / properties / passages / items / properties / datasetVersion / description

Previous value: -"Dataset version at retrieval time; cite this"New value: +"Collection version at retrieval time; use documentVersion for an exact document citation"

changedOutput schema / required

Previous value: -[
-  "query",
-  "passages"
-]New value: +[
+  "query",
+  "passages",
+  "assessment",
+  "guidance"
+]

First observedSep 21, 2026

TDQS

A4.5/5.0

Behavior5/5

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

The description adds critical behavioral information that goes beyond the structured hints, specifically stating that 'Similarity scores rank matches and do not measure factual confidence.' This manages user expectations regarding the accuracy of the retrieved data, which is vital for an LLM agent.

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

Conciseness5/5

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

The two sentences are extremely efficient. The first defines the scope and target resource, while the second immediately provides a critical caveat about the output quality, avoiding unnecessary fluff.

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?

The description covers the resource (passages), the source (Koskamo datasets), the action (search/retrieve), and the inherent limitation (similarity vs. factual confidence). Given the strong annotations and defined schema, this context is functionally complete for a retrieval tool.

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

Parameters3/5

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

The schema coverage is moderate (67%). The description does not elaborate on the semantics of the parameters (e.g., why 'query' should be a question vs. a keyword, or how 'top_k' affects performance). It adds functional context rather than clarifying schema data.

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?

The description clearly states the tool performs a search ('Search Koskamo datasets') to retrieve specific entities ('candidate passages') with unique attributes ('source citations and document versions'). This is distinct from the siblings which deal with listing, registering, or reporting facts.

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?

The tool provides clear usage context by specifying that it searches within 'Koskamo datasets' and that it uses similarity scoring for ranking. While it doesn't explicitly state 'do not use this if you want to list datasets,' the function's purpose highly distinguishes it from sibling tools like koskamo_list_datasets.

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

All Koskamo tools Inspect

Try in Browser

Context All Your Agents Can ShareAgents share what they learn, hand off work, and hold it when you switch models. Start with 5 GB free.fulcradynamics.comAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?