Skip to main content
Glama

seodraft

by app.seodraft
OverviewDiscussionsAdmin

Add evidence

add_evidence
Idempotent

Add pieces to the evidence bank (what only this business knows): a measured number, a named case, a mistake made, a credential, a screenshot to take. Every article draws at least three accepted pieces from it. saidByHuman is true only when transcribing something the human said in this conversation; anything inferred, measured or fetched is false and waits for the human's approval in the app. A true piece is stored accepted with source "conversation"; a false piece is stored pending, and citing it in brief.evidence fails the rules as evidence-unaccepted until the human accepts it. weakFirsthand: true marks a piece that reads as a general claim anyone could repeat rather than something this business knows; it never blocks the write. Deduped on the text, so re-sending a piece never creates a second row and never resets a verdict the human already gave: duplicates names those and returns the row they collapsed onto. pendingReview is the number of pieces in the whole bank waiting for the user's review.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
itemsYes
workspaceIdNoThe workspace this call acts on, as listed by list_workspaces. Left out, the call uses the workspace the account is in right now.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • addedInput schema / properties / workspaceId
      Added value: +{
+  "description": "The workspace this call acts on, as listed by list_workspaces. Left out, the call uses the workspace the account is in right now.",
+  "type": "string"
+}
  2. First observed

TDQS

A4.2/5.0
Behavior5/5

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

Less explicit than annotations alone: it explains the lifecycle and side effects that annotations don't cover. It details saidByHuman semantics (true only for transcription; inferred/measured/fetched false and waits for approval), storage outcomes (true becomes accepted with source 'conversation'; false becomes pending and blocks brief.evidence until accepted), weakFirsthand semantics (never blocks), deduplication on text (never resets a human verdict, duplicates named in response), and pendingReview meaning. This adds rich behavioral context beyond the provided hints such as idempotentHint, destructiveHint, readOnlyHint, openWorldHint.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is dense and front-loaded with the core purpose, but it is a long run-on paragraph mixing multiple concerns (purpose, saidByHuman logic, storage outcomes, weakFirsthand, deduplication, pendingReview) without clear structure. It is not unduly verbose for the complexity, but readability and scanability could be improved with paragraph breaks or bullets. Every sentence does carry distinct information.

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 2-parameter mutation tool with no output schema and 50% schema coverage, the description is quite complete: it covers purpose, approval workflow, deduplication behavior, and response fields. It leaves some gaps around the kind enum and source parameter semantics, and doesn't explicitly say the tool is non-destructive/openWorld, but annotations cover safety and the description covers the nuanced business logic that annotations cannot.

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 50%, so half the parameters (items and its nested fields) are only enumerated as a shape in the schema, not explained in the description. The description explains the semantics of saidByHuman, weakFirsthand (though not in the schema at the top level), source 'conversation', and deduplication on text, adding value. But it does not fully compensate for the coverage gap, nor explain kind enum or source field meaning. Baseline 3 is appropriate.

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?

The description states a specific verb (add) and resource (evidence bank) and immediately qualifies the scope with a parenthetical definition: 'what only this business knows' followed by concrete examples (a measured number, a named case, a mistake made, a credential, a screenshot to take). This distinguishes it from siblings like add_topics or list_evidence, and an agent can tell what belongs in this tool without opening the schema.

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 description gives clear context: it is the source pool that articles draw from ('Every article draws at least three accepted pieces from it'), implying when to add. However, it does not explicitly contrast with siblings like add_topics or list_evidence, or state when not to use this tool. Usage is strongly implied but not fully routed against alternatives.

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

All seodraft toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources