Skip to main content
Glama
zb-ss

Servonaut

by zb-ss
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

remember_server_finding

Save hard-won, non-obvious server discoveries invisible to fresh probes—such as hidden dependencies or load-only bugs—so future agent sessions recall them.

Instructions

Persist a hard-won, non-obvious discovery (quirk, gotcha, root-cause, constraint) about an instance that is NOT visible in a fresh probe — e.g. a misconfigured cron, a hidden dependency, a port blocked by an upstream policy, a bug triggered only under load. Saved locally immediately and queued for end-to-end encrypted sync. The title is the searchable recall key — keep it short and specific. Returns {finding_id, instance_id, title, auto_inject, superseded, secret_warning}. auto_inject=true means the title will be surfaced automatically in future context (confidence >= threshold); false = recall-only.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
bodyYesFull finding text, evidence, and context (≤8000 chars).
tagsNoOptional tags for filtering (max 12, lowercased).
titleYesShort, searchable title for this finding (≤200 chars). This is the primary recall key — make it specific.
confidenceNoConfidence score 0.0–1.0. Values >= threshold (default 0.6) cause the title to be auto-injected into future context; lower values are recall-only.
instance_idYesInstance ID or name. '<account>/<name>' picks the server of one provider account; custom servers are 'custom/<name>'.
supersede_idNoID of an existing finding this corrects or replaces. The old finding is marked superseded; pass the finding_id returned by a previous remember_server_finding call.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changedv2.28.0
    • changedInput schema / properties / instance_id / description
      Previous value: -"Instance ID, name, or custom-server name."New value: +"Instance ID or name. '<account>/<name>' picks the server of one provider account; custom servers are 'custom/<name>'."
  2. Addedv2.24.0
  3. Removedv2.23.8
  4. First observedv1.0.0

TDQS

A4.3/5.0
Behavior4/5

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

With no annotations, the description carries the full burden and does substantial work: it discloses immediate local persistence, queued end-to-end encrypted sync, the auto_inject threshold behavior tied to confidence, and that superseding marks the old finding. It does not mention permission requirements, rate limits, or failure modes, which keeps it short of 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?

Purpose is front-loaded in the first sentence, followed by qualifying examples, persistence semantics, and the return contract. The parenthetical example list is somewhat dense, but each sentence adds distinct information rather than restating the name or schema.

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?

Although there is no output schema, the description enumerates the return object's fields and explains auto_inject and superseded, which is what an agent needs to reason about subsequent calls. Remaining gaps (permissions, error behavior, whether body text is also searchable) are minor for a write-and-queue tool.

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 description coverage is 100%, so the baseline is 3. The description goes beyond per-parameter docs by linking title to the searchable recall key, tying confidence to the returned auto_inject flag, and explaining that supersede_id consumes the finding_id returned by a prior call — a cross-parameter contract the schema alone does not spell out.

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 (persist) and a tightly scoped resource (a hard-won, non-obvious discovery about an instance that is NOT visible in a fresh probe), with concrete examples of what qualifies. This clearly distinguishes it from the read-side sibling recall_server_findings and from memory-building tools like build_server_memory.

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?

The phrase 'NOT visible in a fresh probe' plus the examples (misconfigured cron, hidden dependency, blocked port, load-only bug) give a clear qualifying test for when to use this tool. However, it never explicitly names recall_server_findings as the retrieval counterpart or states when not to write a finding, so the when-not side is left to inference.

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