mcp-stashbox

Get one tag

get_tag

Read-only

Read one catalogue tag by its instance:uuid identifier and return a merged card with source attribution, conflict handling, and optional source filtering.

Instructions

Read one tag from the catalogue its identifier names, written instance:uuid. The answer is one card, read on every catalogue that holds the record and reached by the link each of them publishes to the same record elsewhere. Every value names the catalogues that said it, and where they disagree the reading nobody preferred is published beside the one that won. Name 'sources' to read one catalogue alone.

Input Schema

Name	Required	Description
`id`	Yes
`prefer`	No	The order the catalogues are preferred in where they disagree on a field. Left out, the registry's own order stands, and every card states the order applied.
`sources`	No

Output Schema

Name	Required	Description	Default
`card`	Yes
`cached`	No

Schema Changelog

Changes observed during successful MCP inspections.

Changed2 schema fields changedv2.1.0Oct 30, 2026
- removedOutput schema / properties / card / properties / counts / additionalProperties / items / properties / value / anyOf
```
Removed value: -[
-  {
-    "type": "number"
-  },
-  {
-    "type": "null"
-  }
-]
```
- addedOutput schema / properties / card / properties / counts / additionalProperties / items / properties / value / type
```
Added value: +[
+  "number",
+  "null"
+]
```

Changed3 schema fields changedv1.0.0Aug 19, 2026

changedInput schema / properties / prefer / description

Previous value: -"The order the catalogues are preferred in where they disagree on a field. Left out, the order the registry declares stands. Every card states the order that was applied."New value: +"The order the catalogues are preferred in where they disagree on a field. Left out, the registry's own order stands, and every card states the order applied."

changedOutput schema / properties / card / properties / held_by / description

Previous value: -"Every catalogue asked, with the identifier the record carries there."New value: +"Every catalogue asked, with the identifier the record carries there. This is where what each of them did with the record is read: the state says which of the three it met, and the reason says why it was not asked or what went wrong."

removedOutput schema / properties / per_source

Removed value: -{
-  "items": {
-    "additionalProperties": false,
-    "properties": {
-      "algorithms_not_searched": {
-        "description": "Fingerprint algorithms its lookup does not search, so they were never put to it.",
-        "items": {
-          "type": "string"
-        },
-        "type": "array"
-      },
-      "count": {
-        "description": "Rows it contributed. Never added to another catalogue's: they index corpora that overlap by an amount none of them publishes.",
-        "type": "number"
-      },
-      "error": {
-        "enum": [
-          "not_found",
-          "invalid_input",
-          "rate_limited",
-          "parse_failure",
-          "network_error",
-          "timeout"
-        ],
-        "type": "string"
-      },
-      "index_total": {
-        "description": "What its own index holds for this question, the rows on this page included.",
-        "type": "number"
-      },
-      "index_total_over_any_word": {
-        "description": "The total counts rows carrying any word of the query, its text index reading them apart.",
-        "type": "boolean"
-      },
-      "moment": {
-        "description": "Which moment failed.",
-        "type": "string"
-      },
-      "name": {
-        "description": "The name that catalogue calls itself.",
-        "type": "string"
-      },
-      "narrowings_naming_no_record": {
-        "description": "Narrowings written only with identifiers another catalogue minted.",
-        "items": {
-          "type": "string"
-        },
-        "type": "array"
-      },
-      "narrowings_not_received": {
-        "description": "Narrowings this catalogue cannot receive. This is the one field that says a catalogue cannot do something.",
-        "items": {
-          "type": "string"
-        },
-        "type": "array"
-      },
-      "narrowings_received_in_part": {
-        "items": {
-          "type": "string"
-        },
-        "type": "array"
-      },
-      "reason": {
-        "description": "Why it was not asked, or what went wrong.",
-        "type": "string"
-      },
-      "records": {
-        "description": "Distinct records behind its rows.",
-        "type": "number"
-      },
-      "skipped": {
-        "description": "Rows it answered with that this client could not read and left out.",
-        "type": "number"
-      },
-      "source": {
-        "description": "The catalogue this row is about.",
-        "type": "string"
-      },
-      "state": {
-        "description": "'answered' looked, and a count of zero means it found nothing. 'failed' could not answer, and states nothing about what it holds. 'absent' was never asked. Only the first is evidence about the world.",
-        "enum": [
-          "answered",
-          "failed",
-          "absent"
-        ],
-        "type": "string"
-      },
-      "unattributed": {
-        "description": "Records it answered with while carrying none of what was asked about.",
-        "type": "number"
-      }
-    },
-    "required": [
-      "source",
-      "state"
-    ],
-    "type": "object"
-  },
-  "type": "array"
-}

Addedv0.9.0Aug 14, 2026

TDQS

B3.2/5.0

Behavior4/5

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

Annotations already declare readOnlyHint and openWorldHint, so the description's added value is the federation model: the card is assembled across every catalogue holding the record, each value is attributed to the catalogues that asserted it, and disagreements expose both the winning and non-preferred readings. That is genuine behavioral context an agent cannot get from the structured fields, though the not-found / partial-failure behavior is unstated.

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

Conciseness3/5

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

Purpose is front-loaded and the four sentences all carry content, but the phrasing is needlessly riddling — "the reading nobody preferred is published beside the one that won" must be decoded, and "instance:uuid" is imprecise where a catalogue prefix is what is actually required.

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

Completeness3/5

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

With an output schema present, return shape need not be explained (though the description does sketch the card). Missing pieces remain: behavior when the id resolves to nothing, which prefix to supply, and how this relates to get_sources, which is directly relevant given the 'sources' parameter overlap.

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?

Schema coverage is only 33%: 'id' has a pattern but no prose and 'sources' has no description at all. The description partially compensates by explaining that naming 'sources' restricts to one catalogue and by gesturing at preference ordering ("the reading nobody preferred"), but it never ties that explicitly to 'prefer' and leaves the id prefix format ambiguous.

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

Purpose3/5

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

"Read one tag from the catalogue its identifier names" states the verb and resource, but the qualifier "written instance:uuid" is cryptic and does not match the schema's actual '<catalogue>:<uuid>' prefixes. Nothing distinguishes it from search_tags or get_sources, so the agent must infer it is the single-record lookup for tags.

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

Usage Guidelines3/5

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

The only guidance is "Name 'sources' to read one catalogue alone," which is a conditional usage hint for one parameter rather than a when-to-use-this-vs-alternatives statement. No mention of search_tags for discovery or get_sources for catalogue metadata, so usage is implied at best.

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

Deploy Server