AIWorkhub

Overview Schema Source Related Servers Discussions Admin

aiworkhub_manager_ai_memory_write

Stores, updates, supersedes, or archives canonical AI memory by key with idempotency and provenance tracking, so agents can keep project knowledge current and auditable.

Instructions

MANAGER WRITE: remember, update, supersede, or archive canonical memory.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "action": "<action>",
      "key": "<key>",
      "idempotency_key": "<idempotency_key>",
      "provenance": "<provenance>"
    },
    "name": "aiworkhub_manager_ai_memory_write"
  }
}

Input Schema

Name	Required	Description	Default
`key`	Yes	Memory key, non-blank, at most 256 bytes; update, supersede and archive need an active memory under it.
`tags`	No	Tag string, at most 1024 bytes; default empty.
`scope`	No	Memory scope, non-blank, at most 64 bytes; default project.	project
`value`	No	Memory text, at most 32 KiB; required except for archive.
`action`	Yes	remember adds a memory for a key with no active one, update changes the active one, supersede stores a new row and marks the old one superseded, archive retires it.
`provenance`	Yes	Non-blank provenance text recorded in the audit row, at most 2048 bytes.
`idempotency_key`	Yes	8-192 characters matching ^[A-Za-z0-9][A-Za-z0-9_.:-]{7,191}$; replaying a key returns the first result instead of writing again.

Output Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

Changed7 schema fields changedv0.12.67Oct 26, 2026

addedInput schema / properties / action / description

Added value: +"remember adds a memory for a key with no active one, update changes the active one, supersede stores a new row and marks the old one superseded, archive retires it."

addedInput schema / properties / idempotency_key / description

Added value: +"8-192 characters matching ^[A-Za-z0-9][A-Za-z0-9_.:-]{7,191}$; replaying a key returns the first result instead of writing again."

addedInput schema / properties / key / description

Added value: +"Memory key, non-blank, at most 256 bytes; update, supersede and archive need an active memory under it."

addedInput schema / properties / provenance / description

Added value: +"Non-blank provenance text recorded in the audit row, at most 2048 bytes."

addedInput schema / properties / scope / description

Added value: +"Memory scope, non-blank, at most 64 bytes; default project."

addedInput schema / properties / tags / description

Added value: +"Tag string, at most 1024 bytes; default empty."

addedInput schema / properties / value / description

Added value: +"Memory text, at most 32 KiB; required except for archive."

First observedv0.1.0Oct 20, 2026

TDQS

B3.2/5.0

Behavior3/5

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

Annotations already declare readOnlyHint=false, destructiveHint=false, and openWorldHint=false, so the agent knows this is a local mutation that isn't meant to be destructive. The description adds 'canonical' as a durability signal and hints at lifecycle transitions via 'supersede' and 'archive', but says nothing about the audit row, idempotency behavior, or that archive/supersede only work on an existing active memory - all of which are only in the schema.

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?

A single front-loaded line with zero filler; the action list arrives immediately. It is efficient, though the terseness means the line is doing almost no work beyond the tool name.

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, 100% schema coverage, and annotations covering the safety profile, the description doesn't need to explain return values or arguments. However, for a 7-parameter mutation tool with four distinct state-changing modes, one line of routing-free prose remains thin - a note on checking for an existing active memory before update/supersede/archive would close the gap.

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 description coverage is 100%, so every parameter - including the action enum semantics, key length limits, and idempotency replay behavior - is already fully documented in structured fields. Baseline 3 is appropriate since the description contributes no additional parameter meaning.

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

Purpose4/5

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

States a specific verb set ('remember, update, supersede, or archive') against a specific resource ('canonical memory'), which cleanly separates it from the read-side siblings like ai_memory_get, ai_memory_search, and ai_memory_review. It stops short of 5 only because 'MANAGER WRITE:' is a generic prefix and it doesn't name an alternative tool.

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

Usage Guidelines2/5

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

There is no when-to-use or when-not-to-use guidance. The description never says to prefer memory_search/get first to check for an existing active memory, nor does it say which of the four actions an agent should reach for outside the action names themselves. All routing logic is left to the schema.

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

Deploy Server