Maha Context Compiler

Compress long documents and RAG inputs into token-budgeted, deduplicated context packs with source-linked provenance. Returns original and compiled token counts, so the saving is verifiable. Net-positive above N = fee / (r x p) tokens: r your reduction, p your input price. Extractive and budget-bound, so check includedPassages before relying on it. Ranks Latin, Cyrillic, Greek, Arabic and CJK; CJK

Evaluate exact retention of 1-32 caller-labelled evidence spans at exactly five ascending token budgets. Returns five source-linked Context Packs, per-run retention metrics, an evidence frontier, stable hashes, and a deterministic receipt digest. Price basis: five $0.01 evaluations. Retention is not truth, accuracy, answer quality, legal compliance, or hallucination prevention. Bodies are not stor

Validate exact UTF-8 SHA-256 hashes for up to 64 public or synthetic passages. No Unicode normalization, source authentication, or semantic validation.

Compare two structured celestial result declarations using versioned rules. Reports mismatched conventions and missing prerequisites. No inferred settings, time/frame conversion, calculation verification or predictive validation.

Read one supported section of The Volcanic Engine as machine-readable licensed prose with edition and section identity. For agents researching this specific book. Not the complete edition, a general book search or permission to redistribute the work.

Build an astronomy or astrology tool with seven classical-body geocentric longitudes at one UTC instant, Tropical and versioned Lahiri Sidereal frames, conventions and a deterministic receipt. Public or synthetic inputs only. No interpretations, forecasts or predictive certification. Request and result bodies are not stored.

Check declared MCP tool contracts before an upgrade. Compare required names and exact input/output schema digests with a supplied server snapshot, protocol version and transport. No server contact, semantic schema-subtyping proof or official conformance certification.

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

Choose a context budget before an agent run. Compare five extractive compilation budgets with token estimates and source coverage. Price basis: five compiler runs, the comparison table and the receipt digest. Uses caller-supplied documents; no answer generation, factual verification or guaranteed token-billing savings.

Read one supported section of The Imagined Life as machine-readable licensed prose with edition and section identity. For agents researching this specific book. Not the complete edition, a general book search or permission to redistribute the work.

Get an inspectable exact solution to a 1-8 variable rational linear system. Returns rank, pivots, particular solution and null space; distinguishes unique, inconsistent and underdetermined systems. No floating-point conditioning or physical-model assessment.

Great-circle angles for bounded caller-supplied right ascension and declination pairs in integer microdegrees; no ephemeris lookup.

Compare up to 128 expected passage bindings with up to 128 retained descriptors by passage ID, parent document, version hash and chunk hash. Flags omissions, substitutions, repeats and ordering gaps. Public/synthetic metadata only; does not inspect prompts or prove what a model saw or used.

Check selected versions of up to 40 declared documents against up to 100 effective-date records at one assessment time. Flags inactive, unknown and missing selections; overlapping active versions remain ambiguous. Public/synthetic metadata only. No inferred amendment graph, legal effect or source validation.

Score up to 64 labelled time intervals and 128 command events: rest false activations, missed trials, wrong commands and first-correct response times. Public or synthetic event metadata only. No raw EEG, participant identifiers, live control, diagnosis or human-enhancement claim.

Compare up to 12 caller-supplied public committee-level totals and filing references at a declared assessment time. Preserves unknown totals, flags stale or amended snapshots and checks cash arithmetic. Caller must attest authorized commercial use and no individual-contributor information. No live FEC fetch, donor data, legal verdict or final-filing certification.

Reconcile up to 80 committee-aggregate or non-personal organizational public/synthetic campaign ledger entries against declared opening and closing balances. Flags duplicate IDs and missing document references. Individual-contributor records are prohibited, including hashed or redacted records. Caller must attest authorized commercial use. No contribution intake, payment execution, filing or legal

Review up to 80 public or synthetic spending records for duplicate IDs, inconsistent budgets, budget overruns and declared payee concentration. Duplicate-ID rows are excluded from totals. Flags are review leads, not fraud or legality findings.

Compare up to 20 automation scenarios using declared setup cost, per-unit costs and volumes. Exact cents, savings and break-even counts; no currency conversion, demand forecast, investment advice or independently measured costs.

Simulate up to 10 sequential assembly scenarios of 100 identical units using one shared crane and declared crews. Factory releases, lift and connection durations are explicit. No lift-capacity, structural, site-safety or optimized scheduling approval.

Compare up to 20 homogeneous rectangular-panel scenarios with fixed orientation, edge margin and kerf. Returns row/column capacity, sheet count and material utilization. Not mixed nesting, toolpaths, optimal packing or fabrication approval.

Compare up to 16 declared observation stations against 64 targets and 16 rectangular obstacles in a bounded 2D scene. Returns visible targets and uncovered IDs. No camera optics, 3D visibility, robot motion or physical inspection claim.

Build a Vimshottari timing display with birth balance, nine mahadashas and nine antardashas of the active mahadasha at a reference instant. Uses actual nakshatra stay time and a 365.2425-day year; includes conventions and receipt. Public or synthetic inputs only. No event prediction or interpretation; bodies are not stored.

Build a reproducible chart view: Tropical and Lahiri Sidereal positions, Sidereal whole-sign houses, mean nodes, geometric aspects and panchanga for one UTC chart. Includes limitations and a deterministic receipt. Public or synthetic inputs only. No interpretation, forecast or predictive certification; bodies are not stored.

Prepare an inspectable context-selection evidence pack from caller-supplied documents and labelled spans. Combines budget comparisons, retention checks and policy-bound receipts. Verification concerns the declared workflow, not factual truth, answer quality or legal compliance.

Multiply or divide two seven-component SI dimension vectors and compare with the expected output dimension. No unit conversion, formula validation, scale-factor inference or physical-model certification.

Check a declared bandwidth and up to 16 tones against the strict Nyquist inequality, with equality reported separately. Computes folded tone frequencies exactly. Assumes uniform sampling, bandlimited input and no jitter; does not verify filters, acquisition hardware or real signal bandwidth.

Check up to 32 JSON argument values against a closed, declared scalar tool contract: types, required fields, code-point length and exact safe-integer bounds. Rounded fractions/underflow are not integers. Returns unknown fields and mismatches. No remote references, arbitrary patterns, tool execution or MCP conformance certification.

Inspect one supplied x402-v2 exact-payment challenge against an explicit expected HTTPS resource, EVM CAIP-2 network, asset, payee and maximum price. Checks both URL declarations, structure and payment-policy fields only; no live probe, payment, facilitator verification or server-authenticity claim.

Classify up to 100 declared measurement intervals against tolerances as pass, fail or hold. Integer millimetres; uncertainty is an absolute interval half-width, not a confidence level. No sensor verification or engineering approval.

Exact rational expected-yield and cost arithmetic for up to 16 scenarios and 12 sequential conditional reject rates. All batch units incur unit cost; rejected units incur an additional penalty. Reports differences from the first scenario and undefined cost per good unit at zero yield. Declared assumptions, not a production forecast.

Check one structured public or synthetic low-stakes experiment plan for timing, comparator, outcome and analysis declarations. Returns deficiencies and an integrity receipt. No registration, trusted timestamp, prediction, power analysis or scientific validation.

Compute binary Brier score and count-weighted ECE for up to 256 declared held-out score/outcome pairs. Supports probability or percent scores, normalized to probabilities. Equal-width bins use exact rational boundaries: lower inclusive, upper exclusive, final bin includes 1. Empty bins report null means/gaps. No clinical validity claim.

Compare up to 64 total extracted spans from at most 16 labeled tools against exact code-point slices of supplied source text and declared UTF-8 hashes. Group identical source spans to expose text disagreements. No OCR/PDF execution, normalization, or completeness claim.

Evaluate up to 64 explicitly fictional strategy/game choices against caller-declared integer utilities. Reports best declared actions, chosen utility, regret and observed-minus-declared error. No hindsight counterfactual outcome inference, financial advice, real-world intelligence assessment or strategy fitness claim.

Compare source coverage on exactly the same declared record cohort across two snapshots. Report per-record gained/lost sources, per-source count changes, and global coverage drift. Reject unknown sources or missing/duplicate records. Coverage does not establish source quality.

Compare up to 24 public/synthetic feature summaries: mean differences, reference-standardized shifts and variance ratios. Explicitly flags zero reference variance. Counts and variance convention are declared, not verified; not a distribution test or clinical inference. No raw EEG, participant identifiers or full participant signals.

Compare up to 80 expected and 80 observed document descriptors against one declared tenant and scope. Flags missing, unexpected and cross-tenant documents; incomplete scope stays undetermined. Public/synthetic metadata only. Not access-control enforcement, corpus completeness or legal verification.

Check up to 80 declared floor areas and assignable ratios, returning exact assignable/support-area totals. Uses one declared area unit and basis-point ratios. No geometry, zoning, code compliance, rentable-area or structural certification.

Verify a bounded civic-ledger-1 public organizational decision suffix against declared checkpoints, including expected-head truncation detection. No Merkle proof or anchor.

Check presence of baseline, primary measure, acceptance threshold and bounded stopping rule; no efficacy or statistical validation.

Compare up to 64 proposed Base-USDC calls with a declared total budget, per-call cap, allowed payees and human-approval requirement. Returns policy mismatches and exact arithmetic totals. Does not authorize, sign or execute payments; caller approval labels are not verified.

Deterministic four-neighbor BFS on a grid of at most 24×24 cells. Returns a shortest obstacle-free route or an explicit unreachable/step-limit result. Toy discrete geometry only; no hardware, collision certification or physical feasibility claim.

Purchase one pinned Markdown section of The Orbital Mind with edition identity, word/byte counts and SHA-256 integrity receipt. Choose a section and confirm the edition digest before paying. Not an EPUB/PDF bundle, medical advice or factual certification. Personal/internal use only; no redistribution, resale, training rights or copyright transfer. No stored-result recovery.

Inspect a declared dependency graph of up to 64 nodes and 128 edges: exact cycle membership, source-root reachability, and deterministic graph digest. Duplicate edges and dangling references are rejected. Hash declarations are not authenticated.

Purchase one pinned Markdown section of The Maha Principle with edition identity, word/byte counts and SHA-256 integrity receipt. Choose a section and confirm the edition digest before paying. Not an EPUB/PDF bundle, medical advice or factual certification. Personal/internal use only; no redistribution, resale, training rights or copyright transfer. No stored-result recovery.

Compare bounded room requirements with supplied topology; report unknown and contradictory declarations, not architectural approval.

Find an exact 0/1 maximum-declared-value allocation for up to 64 evidence items and a 0–4096 integer token budget. Deterministic ID-based tie handling; items are indivisible. Optimality applies only to the additive caller-declared values, not relevance or answer quality.

Check up to 64 workflow steps for duplicate IDs, missing dependencies and cycles. Returns a deterministic topological order only when valid. No scheduling, tool execution, access approval or guarantee that the workflow will succeed.

Fit y = intercept + slope*x to 2–64 declared pairs using positive weights and exact rational normal equations. Returns residuals and weighted SSE rounded to 12 decimal places, plus finite float64 RMSE. Rejects degenerate X values; weights are not verified uncertainties. Not field certification.