Skip to main content
Glama
putervision

agent-reasoning-mcp

by putervision

manage_beliefs

Destructive

Maintain dynamic agent beliefs with TTL expiration, confidence decay, and category filtering. Update, query, expire, or reconcile structured belief states for AI reasoning.

Instructions

Maintain structured belief state with TTL expiration sweeps, exponential confidence decay, and category filtering (actions: update, query, expire, reconcile, reconcile_spatial). Use manage_beliefs instead of query_knowledge when managing dynamic agent epistemic state rather than static heuristic patterns.

Returns belief record, query matches, expired belief count, or reconciliation report.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
actionYesBelief operation: update, query, expire, reconcile, reconcile_spatial
objectNoBelief value / state payload
sourceNoBelief provenance source
projectNoTarget project slug
subjectNoBelief subject (e.g. "north_gate", "enemy_patrol")
categoryNoBelief category
belief_idNoBelief ID for specific lookup
predicateNoPredicate relationship (e.g. "is_locked", "status")
confidenceNoConfidence score (0.0 to 1.0)
decay_rateNoExponential decay rate lambda per hour
expires_atNoISO-8601 expiration timestamp
novel_entitiesNoNovel entities for reconcile_spatial action
matched_entitiesNoMatched entities for reconcile_spatial action
missing_entitiesNoMissing entities for reconcile_spatial action
client_request_idNoIdempotency key
unexpected_entitiesNoUnexpected entities for reconcile_spatial action

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed6 schema fields changedv0.4.0
    • changedInput schema / properties / action / description
      Previous value: -"Belief operation: update, query, expire, reconcile"New value: +"Belief operation: update, query, expire, reconcile, reconcile_spatial"
    • changedInput schema / properties / action / enum
      Previous value: -[
      -  "update",
      -  "query",
      -  "expire",
      -  "reconcile"
      -]New value: +[
      +  "update",
      +  "query",
      +  "expire",
      +  "reconcile",
      +  "reconcile_spatial"
      +]
    • addedInput schema / properties / matched_entities
      Added value: +{
      +  "description": "Matched entities for reconcile_spatial action",
      +  "items": {},
      +  "type": "array"
      +}
    • addedInput schema / properties / missing_entities
      Added value: +{
      +  "description": "Missing entities for reconcile_spatial action",
      +  "items": {},
      +  "type": "array"
      +}
    • addedInput schema / properties / novel_entities
      Added value: +{
      +  "description": "Novel entities for reconcile_spatial action",
      +  "items": {},
      +  "type": "array"
      +}
    • addedInput schema / properties / unexpected_entities
      Added value: +{
      +  "description": "Unexpected entities for reconcile_spatial action",
      +  "items": {},
      +  "type": "array"
      +}
  2. Changed2 schema fields changedv0.3.1
    • changedInput schema / properties / action / description
      Previous value: -"Belief operation"New value: +"Belief operation: update, query, expire, reconcile"
    • addedInput schema / properties / action / enum
      Added value: +[
      +  "update",
      +  "query",
      +  "expire",
      +  "reconcile"
      +]
  3. Changed15 schema fields changedv0.2.1
    • removedInput schema / $schema
      Removed value: -"http://json-schema.org/draft-07/schema#"
    • removedInput schema / additionalProperties
      Removed value: -false
    • addedInput schema / properties / action / description
      Added value: +"Belief operation"
    • removedInput schema / properties / action / enum
      Removed value: -[
      -  "update",
      -  "query",
      -  "expire",
      -  "reconcile"
      -]
    • addedInput schema / properties / belief_id / description
      Added value: +"Belief ID for specific lookup"
    • addedInput schema / properties / category / description
      Added value: +"Belief category"
    • addedInput schema / properties / client_request_id / description
      Added value: +"Idempotency key"
    • addedInput schema / properties / confidence / description
      Added value: +"Confidence score (0.0 to 1.0)"
    • addedInput schema / properties / decay_rate / description
      Added value: +"Exponential decay rate lambda per hour"
    • addedInput schema / properties / expires_at / description
      Added value: +"ISO-8601 expiration timestamp"
    • addedInput schema / properties / object / description
      Added value: +"Belief value / state payload"
    • addedInput schema / properties / predicate / description
      Added value: +"Predicate relationship (e.g. \"is_locked\", \"status\")"
    • addedInput schema / properties / project / description
      Added value: +"Target project slug"
    • addedInput schema / properties / source / description
      Added value: +"Belief provenance source"
    • addedInput schema / properties / subject / description
      Added value: +"Belief subject (e.g. \"north_gate\", \"enemy_patrol\")"
  4. First observedv0.1.2

TDQS

A4.3/5.0
Behavior4/5

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

Annotations already declare destructiveHint=true and idempotentHint=false, and the description adds real behavioral context beyond them: TTL expiration sweeps, exponential confidence decay, and category filtering, plus the shape of the return value. It still doesn't say what the destructive expire path actually removes or how update overwrites prior state, so it falls short of full disclosure.

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

Conciseness4/5

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

Front-loads the capability and the sibling comparison, then closes with the return shapes. Two focused sentences with little waste; the return sentence earns its place because there is no output schema.

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 16-parameter, five-action mutation tool with no output schema, the description compensates by listing return values (belief record, query matches, expired count, reconciliation report) and the action set. The remaining gap is that it never indicates which parameters are required by which action, which matters at this parameter count.

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 all 16 parameters are already documented in the schema and the baseline is 3. The description adds the action vocabulary but no per-action parameter mapping or format detail (e.g., which fields reconcile_spatial requires) 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 ("Maintain") and resource ("structured belief state") and enumerates the five supported actions, so the scope is unambiguous. It also explicitly distinguishes itself from the sibling query_knowledge, letting an agent route 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?

Explicitly names the alternative tool (query_knowledge) and the condition that selects this one instead (dynamic agent epistemic state vs static heuristic patterns). That is a direct when-to-use/when-not-to-use rule rather than implied guidance.

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