Skip to main content
Glama
PseudoLife-MCP

PseudoLife-MCP

Official

memory_recall

Read-only

Retrieve answers to relational questions by following knowledge-graph links across multiple hops when single-shot search cannot chain evidence.

Instructions

Multi-hop retrieval over the knowledge graph, for RELATIONAL questions whose answer is reached by following links — "what does X ultimately run on?", "how does A reach C?" — which single-shot memory_search can't chain. Read-only. low_confidence: true means no seed entity matched — fall back to memory_search.

Returns: {seeds, entities, edges, paths, texts, iterations}. entities/edges/texts are capped with a per-hop reservation, so a hub seed's own 1-hop ring can't crowd out the deeper hops the walk exists to reach; edges prefers links between surviving entities; each entity's facts is capped too. A fact carrying re_verify stands on a memory that has since been corrected — the value still stands, but check it before acting. A seed entity's constraint facts come first, marked pinned. truncated: true (with searches_issued) means a search ceiling or time budget stopped the walk early: some re-queries, and possibly deeper hops, were skipped, so supporting texts and deeper entities may be missing — narrow the question or lower hops. Details: docs/guide/retrieval.md.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "query": "<query>"
    },
    "name": "memory_recall"
  }
}

Input Schema

NameRequiredDescriptionDefault
hopsNoMax graph hops. Clamped to 1..5. When omitted, uses memory.recall.default_hops.
queryYesNatural-language relational question; its top hits seed the graph walk.
top_kNoPositive integer. Bounds only the SEED search (the initial hits that name the walk's start entities), not the result, which is capped separately; up to 3 ``texts`` slots go to seed hits. When omitted, uses memory.recall.default_top_k.
verboseNoFull fact/edge provenance and untruncated texts. Default facts are ``{attribute, value}``, edges ``{src, relation, dst}``, texts truncated to a preview.

Output Schema

NameRequiredDescriptionDefault

No arguments

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed4 schema fields changedv0.22.0
    • changedInput schema / properties / hops / default
      Previous value: -3New value: +null
    • changedInput schema / properties / hops / description
      Previous value: -"Max graph hops. Clamped to 1..5."New value: +"Max graph hops. Clamped to 1..5. When omitted, uses memory.recall.default_hops."
    • changedInput schema / properties / top_k / default
      Previous value: -5New value: +null
    • changedInput schema / properties / top_k / description
      Previous value: -"Positive integer. Bounds only the SEED search (the initial hits that name the walk's start entities), not the result, which is capped separately; up to 3 ``texts`` slots go to seed hits."New value: +"Positive integer. Bounds only the SEED search (the initial hits that name the walk's start entities), not the result, which is capped separately; up to 3 ``texts`` slots go to seed hits. When omitted, uses memory.recall.default_top_k."
  2. Changed3 schema fields changedv0.18.1
    • addedInput schema / additionalProperties
      Added value: +false
    • changedInput schema / properties / top_k / description
      Previous value: -"Bounds only the SEED search (the initial hits that name the walk's start entities), not the result, which is capped separately; up to 3 ``texts`` slots go to seed hits."New value: +"Positive integer. Bounds only the SEED search (the initial hits that name the walk's start entities), not the result, which is capped separately; up to 3 ``texts`` slots go to seed hits."
    • addedInput schema / properties / top_k / minimum
      Added value: +1
  3. Changed4 schema fields changedv0.15.0
    • addedInput schema / properties / hops / description
      Added value: +"Max graph hops. Clamped to 1..5."
    • addedInput schema / properties / query / description
      Added value: +"Natural-language relational question; its top hits seed the graph walk."
    • addedInput schema / properties / top_k / description
      Added value: +"Bounds only the SEED search (the initial hits that name the walk's start entities), not the result, which is capped separately; up to 3 ``texts`` slots go to seed hits."
    • addedInput schema / properties / verbose / description
      Added value: +"Full fact/edge provenance and untruncated texts. Default facts are ``{attribute, value}``, edges ``{src, relation, dst}``, texts truncated to a preview."
  4. First observedv0.11.0

TDQS

A4.5/5.0
Behavior5/5

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

Annotations already cover the safety profile (readOnlyHint=true, destructiveHint=false), and the description adds substantial context beyond them: low_confidence semantics, re_verify facts resting on superseded memories, pinned constraint facts ranked first, per-hop reservation capping for entities/edges/texts, and truncated/searches_issued signaling partial walks. These are non-obvious behaviors an agent could not infer from structured fields.

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?

Purpose is front-loaded and the fallback rule follows immediately, which is good structure. However, the block that enumerates return fields ({seeds, entities, edges, paths, texts, iterations}) and capping rules duplicates the existing output schema, and the paragraph is dense enough that some sentences do not earn their place independently.

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

Completeness5/5

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

Given a complex multi-hop tool with an output schema, the description still supplies the decision-critical context: when to prefer it over memory_search, how to interpret low_confidence, re_verify, pinned, and truncated, and how to recover from a partial walk. There is no material gap for an agent calling this correctly.

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 schema already explains hops, top_k seeding semantics, and verbose provenance. The description only adds marginal meaning by tying 'lower hops' to the truncation remedy; it does not elaborate on how top_k or verbose interact with results beyond what the schema states.

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 and resource ('multi-hop retrieval over the knowledge graph') and scopes it to RELATIONAL questions with concrete examples ('what does X ultimately run on?', 'how does A reach C?'). It also names the sibling it is not ('single-shot memory_search can't chain'), so an agent can distinguish it from memory_search without opening either schema.

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

Usage Guidelines5/5

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

Explicit when-to-use (relational questions requiring chained links), implicit when-not (single-shot queries belong to memory_search), and an explicit fallback rule: low_confidence: true means no seed matched, so use memory_search. It even routes recovery for truncated results ('narrow the question or lower hops').

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