Skip to main content
Glama

kit-memory-demo

by io.github.PeterAtPondo

Saving this conversation

capture

Extract candidate memories from a session transcript and save them as drafts for later review. Use near the end of a meaningful session when the transcript contains decisions, gotchas, or reusable context.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
sourceNoSurface slug for provenance. Defaults to this MCP server's detected surface, e.g. claude-code, codex, cursor, or vscode.
projectNoProject to tag drafts with (resolved against registry)
max_draftsNoCap on candidates Sonnet extracts (default 5)
session_idNoSession identifier for provenance
transcriptYesConversation transcript. Plaintext with [user]/[assistant] speaker markers works; JSONL from Claude Code is auto-detected.
contributor_modelNoWhich model produced this transcript (e.g. 'gpt-4o', 'gemini-2.5-pro', 'claude-sonnet-4', 'kit'). Written into each draft's written_by for attribution.

Schema Changelog

Changes observed during successful MCP inspections. Dates show when Glama detected each change.

  1. Added
  2. Removed
  3. Changed7 schema fields changed
    • addedInput schema / properties / contributor_model / description
      Added value: +"Which model produced this transcript (e.g. 'gpt-4o', 'gemini-2.5-pro', 'claude-sonnet-4', 'kit'). Written into each draft's written_by for attribution."
    • addedInput schema / properties / max_drafts / default
      Added value: +5
    • changedInput schema / properties / max_drafts / description
      Previous value: -"Default 5."New value: +"Cap on candidates Sonnet extracts (default 5)"
    • addedInput schema / properties / project / description
      Added value: +"Project to tag drafts with (resolved against registry)"
    • addedInput schema / properties / session_id / description
      Added value: +"Session identifier for provenance"
    • changedInput schema / properties / source / description
      Previous value: -"Surface label such as cursor, vscode-copilot, claude-code."New value: +"Surface slug for provenance. Defaults to this MCP server's detected surface, e.g. claude-code, codex, cursor, or vscode."
    • changedInput schema / properties / transcript / description
      Previous value: -"Conversation transcript. Plaintext with [user]/[assistant] markers works."New value: +"Conversation transcript. Plaintext with [user]/[assistant] speaker markers works; JSONL from Claude Code is auto-detected."
  4. First observed

TDQS

A3.8/5.0
Behavior3/5

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

Annotations only provide a title, so the description carries the full burden for behavioral disclosure. It states that the tool extracts memories and saves drafts, indicating a write operation. However, it does not explain whether drafts are persistent, how to review them, or any side effects (e.g., does it consume the transcript?). The description is adequate but not rich in behavioral detail.

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

Conciseness5/5

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

Two sentences, zero wasted words. The first sentence states the core purpose, the second provides usage guidance. Information is front-loaded and every sentence earns its place. Ideal structure for an agent to quickly parse.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The tool has 6 parameters (1 required) and no output schema. The description covers purpose and usage but omits crucial details: What does the tool return? Are drafts accessible immediately? Are there any error conditions or prerequisites? The agent is left guessing about the outcome and next steps. Given the absence of an output schema, the description should compensate with at least a hint of the return value or side effects.

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 the baseline is 3. The description does not add parameter-specific information beyond what the schema provides. The schema descriptions are already clear and complete (e.g., 'transcript' format, 'source' default, 'max_drafts' cap). The description adds no extra value for parameters, maintaining the baseline.

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 uses a specific verb-resource combination: 'Extract candidate memories from a session transcript and save them as drafts.' This clearly distinguishes 'capture' from sibling tools like 'remember' (direct storage), 'note' (manual note-taking), and 'recall' (retrieval). The purpose is unambiguous and unique.

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 explicitly advises when to use the tool: 'Use near the end of a meaningful session when the transcript contains decisions, gotchas, or reusable context.' This provides clear situational guidance. It does not explicitly mention when not to use it or name alternatives, but the positive conditions are sufficient for an agent to decide contextually.

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

All kit-memory-demo toolsInspect
Try in Browser
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A3.7/5.0
Disambiguation4/5

Most tools have distinct purposes like listing areas, capturing drafts, committing memories, and searching. However, the 'note' and 'handoff' tools overlap significantly because 'note' can be used with kind=handoff for the same scenarios that 'handoff' is designed for, potentially confusing an agent on which to use.

Naming Consistency4/5

Tool names are all single-word, lowercase, and descriptive. Most are verbs (capture, commit, focus, etc.), but 'areas' and 'presence' are nouns, introducing a minor inconsistency. Overall, the pattern is simple and readable.

Tool Count5/5

With 10 tools covering listing, capturing, committing, searching, retrieving, and setting context, the count is well-scoped for a memory/knowledge management server. Each tool serves a clear purpose and fits within the domain.

Completeness2/5

The tool surface lacks update and delete operations for memories, which are essential for a complete lifecycle. There is no way to modify or remove committed memories, and no tool to list all memories (only semantic search). This creates significant gaps that will likely cause agent failures.

Resources