Skip to main content
Glama
CoderDayton

verifiable-thinking-mcp

scratchpad

Structure reasoning with verification, detect cognitive traps, and self-challenge to improve problem-solving accuracy.

Instructions

Structured reasoning w/verification, trap detection, self-challenge. []=optional

OPS (required: operation=): step thought= [question=1st] [confidence=] [verify=] [domain=math|logic|code|general] [compress=true]→add step. Auto-verifies when chain >3 steps. complete [final_answer=] [summary=]→finalize+spot-check revise target_step= thought= [reason=]→fix step branch thought= [from_step=] [hypothesis=] [success_criteria=]→fork path navigate view=history|branches|step|path [step_id=] [limit=10]→inspect augment text= [store_as_step=false]→compute+inject math results hint [expression=] [reveal_count=] [cumulative=true] [reset=false]→progressive hints (auto-continues) mistakes text=→check algebraic errors spot_check question= answer=→check for common reasoning traps challenge [target_claim=] [challenge_type=all]→adversarial self-check override failed_step= [reason=]→force-commit failed step

DEFAULTS: session_id=auto confidence_threshold=0.8 token_budget=3000 augment_compute=true compress=true

STATUS→ACTION: continue→add steps | threshold_reached→complete or verify | review→use reconsideration.suggested_revise | verification_failed→revise|branch|override | budget_exhausted→complete or new session

FLOW: 1.step(question="...",thought="...")→primes trap detection for the question 2.step(thought="...")×N→auto-verify, auto-compress, confidence-drift detection, consistency checks 3.[optional]challenge()→adversarial self-check of claims 4.complete(final_answer="...")→auto spot-check against common traps 5.if status=review→revise per reconsideration.suggested_revise

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
operationYesOperation to perform
confidence_thresholdNoChain confidence threshold to suggest completion (default: 0.8)
token_budgetNoMax tokens before auto-compressing new steps (default: 3000)
warn_at_tokensNoWarn when cumulative session tokens exceed this threshold (soft limit, cost control)
hard_limit_tokensNoHard stop when cumulative session tokens exceed this threshold. Returns budget_exhausted status and blocks further operations.
thoughtNoCurrent reasoning/analysis (step/branch/revise)
purposeNoStep category
outcomeNoResult or conclusion from this step
confidenceNoConfidence in this step (0-1). Contributes to chain average.
contextNoPrior context or findings
verifyNoRun domain verification. Auto-enabled for chains >3 steps. Set to false to disable.
domainNo
local_computeNoTry local compute for math (default: false)
augment_computeNoAuto-inject computed values into thought (default: true)
compressNoCompress thought before storing (default: true)
compression_queryNoQuery for context-aware compression
max_step_tokensNoMax tokens for this step. Rejects if exceeded (default: no limit)
force_largeNoAllow step even if it exceeds max_step_tokens (default: false)
preconditionsNoAssumptions that MUST be true for this step (e.g., 'x > 0', 'file exists')
viewNoWhat to view: history (all steps), branches (list), step (specific), path (lineage)
step_idNoStep number to view
branch_idNoFilter history by branch
limitNoMax steps to return (default: 10)
from_stepNoStep to branch from (default: current)
branch_nameNoHuman-readable branch name
hypothesisNoFalsifiable hypothesis this branch will test (e.g., 'Assume X is prime')
success_criteriaNoWhat observation proves/disproves this hypothesis
target_stepNoStep number to revise
reasonNoWhy revising this step / Why overriding verification
summaryNoFinal summary/conclusion
final_answerNoThe answer/result
questionNoOriginal question. On step: enables trap priming and stores for auto spot-check. On complete: enables spot-check.
textNoText containing math expressions to compute and inject (augment/mistakes)
system_contextNoSystem prompt context for domain filtering
store_as_stepNoStore augmented result as a reasoning step (default: false)
acknowledgeNoConfirm you understand verification failed but want to proceed
failed_stepNoStep number that failed verification
expressionNoMath expression to simplify. Omit to continue from previous hint in session.
reveal_countNoNumber of steps to reveal. Omit to auto-increment when continuing.
cumulativeNoShow all steps up to reveal_count (true) or just the nth step (false). Default: true
resetNoReset hint state and start from beginning (default: false)
answerNoThe proposed answer to check for trap patterns
challenge_typeNoType of challenge to generate (default: all)
target_claimNoSpecific claim to challenge (optional - if omitted, extracts claims from steps)

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/CoderDayton/verifiable-thinking-mcp'

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