Skip to main content
Glama

Grounded Cognitive Triangulation & Low-Confidence Escalator

kilo_triangulate_research

Compare architectural options with SQLite memory, GitHub grounding, and ToT trade-offs; persist reasoning before code changes and escalate low-confidence research.

Instructions

Execute Triangulated Cognitive Synthesis: Combines Internal SQLite Memory, External GitHub Grounding, 3-Option ToT DAG Trade-Offs, and Low-Confidence Research Escalation. Persists reasoning atomically into SQLite before code modification.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
formatNomarkdown
sessionIdYesActive Kilo-Kit session ID
dagOptionsYesAt least 2 (ideally 3) competing architectural options to compare
chosenOptionYesThe selected option name with rationale
confidenceScoreYesConfidence score between 0.0 and 1.0 (if < 0.70, triggers research escalation)
taskDescriptionYesThe architectural decision, feature spec, or bug to analyze
researchFindingsNoSynthesized findings from research subagent if research escalation was triggered
decision_narrationNoMANDATORY INTER-TOOL NARRATION: State what you just concluded, verified, or analyzed before calling this tool. Appears in terminal tool badge for full user observability.
adversarialRiskScoreNoAdversarial risk score from red-team grilling
internalMemoryLearnedNoKey facts, reflections, or past pitfalls retrieved from SQLite
next_action_narrationNoMANDATORY: State what this tool will execute and what immediate action follows.
externalGroundingPatternsNoStandard patterns, official docs, or 10k+ stars GitHub implementations
requiresResearchEscalationNoExplicitly request deep subagent research fallback

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

B3.1/5.0
Behavior3/5

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

Annotations declare readOnlyHint=false, idempotentHint=false, destructiveHint=false. The description adds useful context beyond that: reasoning is persisted atomically into SQLite and this occurs before code modification, which signals a stateful write with sequencing constraints. It stops short of saying what gets written, whether repeat calls duplicate state, or what the escalation does in practice.

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?

Two sentences, front-loaded with the action verb, so it is compact. However, the second sentence is packed with opaque capitalized jargon ('Grounded Cognitive Triangulation', 'Triangulated Cognitive Synthesis') that consumes space without adding precision.

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

Completeness3/5

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

For a 13-parameter, non-idempotent write tool with no output schema, the description covers the high-level purpose and the persistence behavior but omits usage routing and expected outcomes. The rich schema compensates for parameter detail, but the agent still lacks guidance on when this tool should be invoked versus its siblings.

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 92%, so the schema already documents nearly every parameter including the confidenceScore escalation threshold and the narration params. The description adds no parameter-level detail beyond what the schema provides, so baseline 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a concrete verb ('Execute Triangulated Cognitive Synthesis') and enumerates the distinct operations it combines (SQLite memory, GitHub grounding, ToT DAG trade-offs, research escalation) plus its persistence step. An agent can grasp what the tool does, though the jargon-heavy phrasing and lack of differentiation from siblings like kilo_think_step or kilo_grill_plan keeps it from a 5.

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

Usage Guidelines2/5

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

The description never states when to choose this tool over alternatives such as kilo_think_step, kilo_grill_plan, or kilo_benchmark_solution. The routing logic (confidenceScore < 0.70 triggers escalation) lives only in the schema, not in the description, and no prerequisites or exclusion conditions are given.

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