Skip to main content
Glama

helped

Provide feedback on whether recalled memories were useful to improve AI memory accuracy. This tool strengthens helpful memories and allows unhelpful ones to decay, creating a learning feedback loop.

Instructions

Mark whether a surfaced memory actually helped.

This feedback loop is what makes NeverOnce learn.
Helpful memories get stronger. Unhelpful ones decay.

Args:
    memory_id: The memory ID (from recall results).
    did_help: True if the memory was useful, False if not.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
memory_idYes
did_helpYes

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
resultYes

Implementation Reference

  • The MCP tool handler for 'helped', which calls the memory object's helped method.
    def helped(memory_id: int, did_help: bool) -> str:
        """Mark whether a surfaced memory actually helped.
    
        This feedback loop is what makes NeverOnce learn.
        Helpful memories get stronger. Unhelpful ones decay.
    
        Args:
            memory_id: The memory ID (from recall results).
            did_help: True if the memory was useful, False if not.
        """
        mem = _get_mem()
        mem.helped(memory_id, did_help)
        return f"Memory #{memory_id} marked as {'helpful' if did_help else 'not helpful'}"
  • The memory manager helper that initiates the effectiveness update.
    def helped(self, memory_id: int, did_help: bool) -> None:
        """Mark whether a surfaced memory actually helped.
    
        This is the feedback loop. Memories that help get stronger.
        Memories that don't get weaker over time.
        """
        self.db.update_effectiveness(memory_id, did_help)
  • The database logic that calculates effectiveness and updates the 'times_helped' count.
    def update_effectiveness(self, memory_id: int, helped: bool):
        mem = self.get(memory_id)
        if not mem:
            return
        surfaced = mem["times_surfaced"]
        helped_count = mem["times_helped"] + (1 if helped else 0)
        effectiveness = helped_count / max(surfaced, 1)
        self.conn.execute(
            """UPDATE memories
               SET times_helped = ?, effectiveness = ?, accessed_at = ?
               WHERE id = ?""",
            (helped_count, effectiveness, _now(), memory_id),
        )
        self.conn.commit()
  • The MCP tool registration decorator for the 'helped' function.
    @mcp.tool()
Behavior4/5

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

With no annotations provided, the description carries the full burden of behavioral disclosure and succeeds well. It explicitly describes the side effects: 'Helpful memories get stronger. Unhelpful ones decay.' This explains the learning mechanism impact, which is critical context for a feedback tool that the raw schema cannot convey.

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?

The description is optimally structured: purpose statement first, behavioral mechanism second (the learning loop), then parameter documentation. Every sentence earns its place—no redundant repetition of the tool name, no filler text. The progression from 'what' to 'why' to 'how' is logical and efficient.

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 the tool's simplicity (2 primitive parameters) and the existence of an output schema (noted in context signals), the description is sufficiently complete. It explains the domain-specific learning behavior (strengthening/decay) that agents need to understand the tool's role in the broader system, without needing to document return values.

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 0% (properties lack descriptions), but the Args section compensates effectively. It adds crucial semantic context that memory_id originates 'from recall results' (establishing the workflow dependency) and clarifies that did_help indicates whether the memory 'was useful' (translating the boolean into domain meaning).

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 action ('Mark whether a surfaced memory actually helped') that clearly identifies the verb and resource. It distinguishes from siblings like 'recall' (retrieval) and 'store' (creation) by specifying this is a feedback operation on already-surfaced memories, and further differentiates from 'correct' or 'forget' by focusing on utility marking rather than content modification or deletion.

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

Usage Guidelines3/5

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

The parameter documentation provides clear contextual guidance by stating memory_id comes 'from recall results,' implying the temporal sequence (use after recall). However, it lacks explicit 'when not to use' guidance or differentiation from potentially similar siblings like 'correct' (which might also modify memory metadata).

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

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/WeberG619/neveronce'

If you have feedback or need assistance with the MCP directory API, please join our Discord server