Skip to main content
Glama

dejo_manage

Destructive

Manage what Dejo remembers about a website: forget sites or specific tasks, save useful facts, and share or borrow learned routes with other agents.

Instructions

Change what Dejo remembers about a website. do=forget removes a site (or one task) for good; do=note keeps a fact that is not a step (needs 2FA, use the finance account); do=share offers a trail to other agents (typed values never leave); do=borrow takes a shared trail.

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": {
      "do": "<do>",
      "site": "<site>"
    },
    "name": "dejo_manage"
  }
}

Input Schema

NameRequiredDescriptionDefault
doYes
siteYes
taskNo
textNothe fact, for note

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.5.0

TDQS

A4/5.0
Behavior4/5

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

Annotations only give destructiveHint=true; the description adds the crucial detail that forget is permanent ('for good') and that share excludes typed values ('typed values never leave'), which is real behavioral context an agent could not infer from the schema. What borrow actually returns and whether share is scoped or global remain unstated, keeping it below a 5.

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?

The scope statement is front-loaded and the four modes are packed into a compact semicolon-delimited list with zero filler. The single long sentence is dense but still parseable; a line-per-mode layout would have been marginally easier to scan.

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 four-mode mutation tool with no output schema, the description covers the destructive mode, the privacy constraint on sharing, and the purpose of each operation. The main residual gap is what an agent receives from borrow (and from share) and whether these operate per-site or across all memory.

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?

With schema coverage at 25%, the description must carry the load, and it largely does: all four 'do' values are explained, 'site' is framed as the website in scope, 'task' appears as the narrower forget target, and 'text' is characterized as a fact rather than a step. The mapping of 'task' to the parameter is only implicit, so it is not a full 5.

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?

Opens with a specific verb and resource ('Change what Dejo remembers about a website') and then enumerates the four operations, so the agent knows exactly what capability the tool exposes. It does not distinguish itself from siblings like dejo_memory, but the mode list makes the tool's scope unambiguous on its own.

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?

Each enum value carries an explicit usage rule: forget for permanent removal, note for non-step facts, share to publish a trail, borrow to consume one. That is effectively a when-to-use guide per mode. It stops short of saying when not to use the tool or when a sibling such as dejo_memory is preferable.

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