Skip to main content
Glama

TeamShared

by io.github.xhad

Memory Skill Feedback

memory_skill_feedback

Propose a skill/playbook rewrite from a human-approved draft vs kept.

Stores the pair as a pending rewrite. The live skill does not change. A human accepts or rejects the proposed diff in the console; accept writes vN+1 through the existing setter. Never auto-overwrites.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
keptNoHuman-kept / live body snapshot. Default: current live body
kindNoskill (atomic how-to) or playbook (composed flow)skill
nameYesSkill or playbook name that ran
noteNoWhy this rewrite, or the human-approved result summary
agentNoOverride agent identity
draftYesProposed new body_md (skill) or steps_md (playbook)
extraNoProposed extras: tool_hints/tags (skill) or tool_recipe/tags (playbook)
originYesWho proposed the rewrite: agent, human, or curator
descriptionNoProposed one-line summary

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault

No arguments

Schema Changelog

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

  1. Added

TDQS

A4.4/5.0
Behavior5/5

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

With no annotations, the description carries the full behavioral burden and does so thoroughly: the live skill is unchanged, the pair is stored as pending, a human must accept or reject in the console, accept writes vN+1, and the tool never auto-overwrites. This gives an agent a clear safety model for a mutating feedback operation.

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?

Four sentences, each earning its place: purpose, storage state, live behavior, human approval flow, and the never-overwrite guarantee. The description is dense but not verbose, with the purpose front-loaded.

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?

Given an output schema exists and all 9 parameters are documented, the description covers the workflow adequately. Minor ambiguity in 'human-approved draft vs kept' and no explicit pointer to memory_skill_set as the setter prevent a 5.

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 each parameter is already documented. The description adds only a light framing of 'draft vs kept' and does not provide additional parameter-level detail beyond the schema, meriting the baseline 3.

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 opens with a specific verb and resource: 'Propose a skill/playbook rewrite from a human-approved draft vs kept.' It also distinguishes itself from direct setters by stating 'The live skill does not change' and 'accept writes vN+1 through the existing setter.'

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?

It clearly conveys when to use this tool: to store a pending rewrite that requires human approval, and it explicitly says 'Never auto-overwrites.' It does not name alternatives like memory_skill_set or memory_playbook_set, but the human-in-the-loop review context makes the intended use unmistakable.

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

All TeamShared 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

B3.4/5.0
Disambiguation4/5

With 104 tools across many domains (memory, work, projects, files, agents, context, strategic, ontology), the use of clear prefixes (memory_, work_, project_, file_, agent_run_, context_) makes most tools distinct. However, there are some potential confusions between memory_session_* vs memory_state_*, and memory_recall vs memory_think vs memory_assemble_context, though descriptions clarify their specific purposes. Aliases like memory_playbook_get for memory_procedure_get are explicit and reduce ambiguity.

Naming Consistency5/5

Tool names follow a highly consistent pattern: prefix_domain_action (e.g., file_create, work_update, memory_recall, agent_run_start). All use snake_case, with verbs consistently placed after the domain prefix. Even less common tools like account_brief and attention_snapshot fit the overall naming scheme, making the set predictable and easy to navigate.

Tool Count3/5

At 104 tools, this is an exceptionally large surface area, far exceeding the 25+ threshold that feels heavy. However, the server covers an extensive domain (organizational memory, work management, project tracking, file sharing, agent orchestration, and strategic planning), which justifies a large count. Still, the sheer number may overwhelm agents, and some tools could be consolidated (e.g., many memory_session_* and memory_state_* variants).

Completeness4/5

The tool surface is remarkably complete for its stated purpose, covering CRUD operations for files, work items, projects, and memory, plus lifecycle management for agents, sessions, and strategic plans. Minor gaps exist (e.g., no direct memory_item_get by ID, no section removal in projects), but agents can work around these using existing tools like memory_recall or work_create with parent_id. Overall, the set minimizes dead ends.

Resources