Skip to main content
Glama
Sign In
Linksee Memory

Linksee Memory

by michielinksee
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

share_draft

Builds a link to a reviewed field-note draft after a SaaS, MCP, or API connection issue is fixed or unresolved, so the user can open, review, and send it.

Instructions

Make a LINK to a reviewed draft of a field note for other agents (KanseiLINK note page). ONLY when the user says they want to share — never call it on your own.

Use after a SaaS / MCP / API connection got stuck and was fixed (or was not). You write the fields; this tool only checks them and builds the URL. Linksee sends nothing: the user opens the link, reviews the filled form, and only their click on that page sends it.

No values, only kinds: write "401 invalid_token", not the token. A field that looks like a key, token, JWT, e-mail, tenant URL (acme.slack.com), IP, local path (C:, /Users/, /home/) or a long hex/base64 run stops the build and returns which field. The whole URL must fit in 2,000 characters (a Japanese character takes 9) — if not, it asks you to shorten; it does not cut.

Returns { url, notice } — show the user both. Linksee records only that a draft was made (no field text).

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": {
      "service": "<service>",
      "task": "<task>",
      "symptom": "<symptom>",
      "tried": "<tried>",
      "result": "<result>",
      "kind": "<kind>"
    },
    "name": "share_draft"
  }
}

Input Schema

NameRequiredDescriptionDefault
kindYesexecution = you ran it · documentation = you read it · hypothesis = untested guess
taskYesWhat you were trying to do (≤300 chars)
triedYesThe steps you tried (≤2,000 chars)
resultYesHow it was solved, or that it was not (≤1,000 chars)
runtimeNoOptional. Which AI wrote this
serviceYesKanseiLINK service_id, lowercase (e.g. freee, slack, google-drive). ^[a-z0-9_.-]{1,160}$
symptomYesThe KIND of error that came back — no values (≤600 chars)
conditionsNoOptional. Tool + version, kind of environment (≤600 chars)

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv0.17.0

TDQS

A4.8/5.0
Behavior5/5

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

No annotations exist, so the description carries the full burden and does: it discloses validation behavior (secret/key/JWT/e-mail/tenant-URL/IP/local-path/hex detection halts the build and names the offending field), a hard 2,000-character URL cap with a multibyte note (a Japanese character takes 9) and that it asks for shortening rather than truncating, the {url, notice} return pair, and the privacy posture (Linksee records only that a draft was made, no field text).

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 purpose and the consent rule in the first sentence, then safety/validation details, then the return contract. Dense but every sentence adds a behavior an agent could otherwise get wrong; the length is defensible for a tool with this much implied policy.

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?

Eight parameters with six required and no output schema, yet the description compensates: it explains that the agent authors the fields while the tool only validates, describes the return shape ({url, notice}) and instructs showing both, and covers the failure modes (blocked field, over-length URL). Nothing essential to a correct first call is missing.

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 per the baseline this is a 3; the description earns an upgrade by adding a semantic rule the schema cannot express — 'no values, only kinds', with the concrete example 'write "401 invalid_token", not the token'. That materially shapes how symptom and result should be written. It does not walk through each of the 8 parameters, so not a 5.

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 artifact: builds a LINK to a reviewed draft of a field note (KanseiLINK note page). It immediately distinguishes itself from the memory-oriented siblings by clarifying that this tool only checks fields and constructs a URL, while the user's click on the page is what actually sends anything.

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

Usage Guidelines5/5

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

Explicit when-to-use — after a SaaS/MCP/API connection got stuck and was fixed (or was not) — and an explicit when-not: only when the user says they want to share, never on the agent's own initiative. That is the strongest form of routing guidance an agent can get.

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

Deploy Server

Other Tools