Maha Context Compiler · www.mahastrategies.com

Compile 1-8 documents into a token-budgeted context pack, then…

Compile 1-8 documents into a token-budgeted context pack, then measure exact retention of 1-32 caller-labelled evidence spans. Returns the pack, source-linked passages, input and output hashes, original and compiled token estimates, tokensSaved, source coverage, and requiredEvidenceRetentionPercent. Retention is exact span matching only: not accuracy, answer quality, verification, or hallucination prevention. Extractive and budget-bound. Request content is not stored.

Ready now4 calls · 2 buyers in 30 days

Input

Request JSON

{
  "budgetMode": "guaranteed",
  "clientRequestId": "req_rag_release_eval_001",
  "documents": [
    {
      "id": "release-notes",
      "text": "Release may proceed after the security owner attaches credential-rotation evidence. The production canary passed in every region.\n\nRoutine notes cover staffing, dashboards, meeting cadence, maintenance calendars, regional handoffs, historical capacity, and documentation formatting.",
      "title": "Release notes"
    },
    {
      "id": "rollback-runbook",
      "text": "Rollback if API errors exceed 2 percent for five minutes or payment failures breach the alert threshold. Restore the last-known-good deployment before reopening traffic.\n\nRoutine notes cover staffing, dashboards, meeting cadence, maintenance calendars, regional handoffs, historical capacity, and documentation formatting.",
      "title": "Rollback runbook"
    }
  ],
  "provenance": "compact",
  "requiredEvidence": [
    {
      "evidenceId": "release-gate",
      "sourceId": "release-notes",
      "text": "Release may proceed after the security owner attaches credential-rotation evidence."
    },
    {
      "evidenceId": "rollback-trigger",
      "sourceId": "rollback-runbook",
      "text": "Rollback if API errors exceed 2 percent for five minutes"
    },
    {
      "evidenceId": "handoff-note",
      "sourceId": "release-notes",
      "text": "regional handoffs, historical capacity, and documentation formatting"
    }
  ],
  "scoring": "bm25",
  "task": "Find the release condition and rollback trigger while removing duplicate operational background.",
  "tokenBudget": 128
}

Sample response

Published by the provider. Your run returns a fresh response.

response15 keys

clientRequestId"req_rag_release_eval_001"

compiledContextStoredfalse

contextPack14 keys

clientRequestId"req_rag_release_eval_001"

context"# Context Pack Task: Find the release condition and rollback tr…"

includedPassages[…]1 item

inputHash"sha256:288dd5cacd92158f9460346ba6e931f6d0d52f2598bd55e748fc30a18…"

metrics{ compiledBytes, compiledEstimatedTokens, duplicatePassagesRemoved, … }9 keys

outputHash"sha256:80a337617475688482476a469f9fa7b489ef3e1ad15b4a9756fdc1aa7…"

packId"ctxpack_d0dc187dd43a403d8aef472de5689408"

retentionBoundaries{ claimVerificationPerformed, completenessGuaranteed, evidenceRetention, … }6 keys

sources[…]1 item

task"Find the release condition and rollback trigger while removing d…"

tokenBudget128

version"0.1.0"

warningCodes[…]2 items

warnings[…]1 item

evaluationId"ctxeval_4f1c8a2b6d5e47c0913a7e2f8b46d5c1"

evidence1 item

0{ evidenceHash, evidenceId, sourceId, … }4 keys

inputHash"sha256:98eae8dca77e0c300502f19226528ef22ee2cb8f05687c3513920f91f…"

metrics12 keys

compiledBytes461

compiledEstimatedTokens83

duplicatePassagesRemoved1

estimatedReductionPercent14.4

originalBytes606

originalEstimatedTokens97

requiredEvidenceCount3

requiredEvidenceRetentionPercent66.7

retainedEvidenceCount2

sourceCount2

sourceCoveragePercent100

tokensSaved14

offerId"deep-context-evaluation"

outputHash"sha256:133714e20e990c929f2fbe4693b719cfe19254a92afd0c46ed376cae4…"

requiredEvidenceTextStoredfalse

retentionBoundaries9 keys

answerQualityAssessedfalse

claimVerificationPerformedfalse

completenessGuaranteedfalse

evidenceRetention"best_effort"

evidenceRetentionMeasurement"exact_span_match"

factualAccuracyAssessedfalse

hallucinationPreventionGuaranteedfalse

selectionType"extractive"

tokenCountType"model_neutral_estimate"

sourceTextStoredfalse

version"0.1.0"

warningCodes3 items

0"model_neutral_token_estimates"

1"extractive_selection_not_verification"

2"exact_span_retention_not_accuracy"

warnings1 item

0"Token counts are model-neutral estimates, not provider tokenizer…"

responseobject, 15 keys

Your agent can do this on its own

Add the MCP URL in Claude Code, Cursor or Codex. Your agent can run Tools within its spending limits, or ask for approval each time. Connect your agent

Connect your agent

Add one MCP URL and your agent can find and run this tool within its spending limits. Choose “Ask each time” if you want to approve every Call.

MCP URL

https://app.aecon.ai/mcp

Terminalbash

claude mcp add --transport http --scope user agentic-economy "https://app.aecon.ai/mcp"

In Claude Code, open /mcp, select agentic-economy, then choose Authenticate.

Track record

Paid calls, last 30 days: 4
Buyers, last 30 days: 2
Last paid call: 2 weeks ago

Reported by the x402 directory.

A$0.015 per call