Kosmodrom

memory_update

Destructive

Updates an agent memory record at the user's request; passed fields are replaced, and a new level, projectID or repositoryID moves it. Authors edit their own records, moderators (memory:moderate) any but user-level ones; an author's edit makes the record personal and unverified again, while a moderator's edit keeps these flags; verified and shared (shared needs verified) are a human moderator's decision. Rejected if the record changed since the given version.

Input Schema

Name	Required	Description
`ID`	Yes	Record ID.
`jobs`	No	Jobs the record is for; at least one for job and request, none for always.
`load`	No	always - delivered with any work, short; job - with the jobs in `jobs` (prevention, silent pitfalls); request - only in the `memory` index, read by ID (symptom diagnostics, narrow areas).
`text`	No	One topic, 1.5-3 thousand characters (max 4,000), in the language of the workspace data: rules and facts with reasons. Author and date are record fields, no "Source" line needed.
`level`	No	New level (moves the record).
`title`	No	"<area> · <situation or symptom>: <what's inside>", up to 200 characters, in the language of the workspace data; agents judge relevance by the title in the memory index.
`shared`	No	Requires memory:moderate: shared with all company agents (true) or personal to the author's (false).
`version`	Yes	Record version the change is based on.
`verified`	No	Requires memory:moderate: the record is verified.
`projectID`	No	New project ID, for level = project.
`productIDs`	No	As when creating a record, the full list; [] - all products of the record's project.
`repositoryID`	No	New repository ID, for level = repository.

Schema Changelog

Changes observed during successful MCP inspections.

Changed2 schema fields changedOct 20, 2026

addedInput schema / properties / productIDs

Added value: +{
+  "description": "As when creating a record, the full list; [] - all products of the record's project.",
+  "items": {
+    "format": "uuid",
+    "type": "string"
+  },
+  "maxItems": 50,
+  "type": "array"
+}

changedInput schema / properties / title / description

Previous value: -"\"<area> · <situation or symptom>: <what's inside>\", up to 200 characters, in the language of the workspace data; agents judge relevance by it, and agents doing other kinds of work see it only up to \": \"."New value: +"\"<area> · <situation or symptom>: <what's inside>\", up to 200 characters, in the language of the workspace data; agents judge relevance by the title in the memory index."

Changed4 schema fields changedOct 8, 2026

removedInput schema / $schema

Removed value: -"http://json-schema.org/draft-07/schema#"

changedInput schema / properties / title / description

Previous value: -"\"<area> · <situation or symptom>: <what's inside>\", up to 200 characters, in the language of the workspace data; agents judge relevance by it."New value: +"\"<area> · <situation or symptom>: <what's inside>\", up to 200 characters, in the language of the workspace data; agents judge relevance by it, and agents doing other kinds of work see it only up to \": \"."

removedInput schema / properties / version / maximum
```
Removed value: -9007199254740991
```
removedInput schema / properties / version / minimum
```
Removed value: --9007199254740991
```

First observedOct 6, 2026

TDQS

A4.5/5.0

Behavior5/5

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

Goes well beyond the annotations (destructiveHint/readOnlyHint) by disclosing replacement semantics, optimistic concurrency ('rejected if the record changed since the given version'), permission rules, and the side effects of an author edit (record becomes personal and unverified again) versus a moderator edit (flags retained). These are exactly the mutation consequences an agent needs before invoking a destructive tool.

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?

Front-loaded with the core action before branching into permission and flag-retention details, and every clause carries meaning. It is dense and semicolon-heavy, but not padded given the tool's complexity.

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

Completeness4/5

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

For a 12-parameter destructive mutation with no output schema, it covers permissions, concurrency, and flag behavior well. It omits only minor operational guidance, such as how the caller should obtain the current version or what a rejection looks like.

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%, so the baseline is 3, but the description adds update semantics the schema does not express: which fields are replaced wholesale and that setting level/projectID/repositoryID relocates the record. It still does not explain interplay such as what happens when productIDs is omitted.

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 specific verb and resource ('Updates an agent memory record') and immediately clarifies the update mode ('passed fields are replaced; a new level, projectID or repositoryID moves it'). This is clearly distinguishable from siblings memory_create, memory_delete, and memory_read.

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?

Gives strong context for when and by whom the tool may be used, including the author-vs-moderator permission split and the memory:moderate requirement. It does not, however, explicitly route the agent to alternatives such as memory_read for fetching the current version before calling this tool.

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

All Kosmodrom tools Inspect

Try in Browser

Context Every Agent Can ShareAgents share what they learn, hand off work, and keep 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?