Skip to main content
Glama

SigRank — AI Operator Benchmarking

by com.signalaf

Rank Paste — Local Token Cascade Calculator

rank_paste
Read-onlyIdempotent

Calculate SigRank cascade metrics from four non-negative token counts without submitting data. Returns Yield, Leverage, Velocity, SNR, and 10xDEV. No data is persisted.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
inputYesTotal input tokens consumed in the session.
outputYesTotal output tokens generated.
cache_readYesTokens read from prompt cache (reused context).
cache_writeYesTokens written to prompt cache (new context stored for reuse).

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
snrNoSignal-to-noise ratio = output / (input + output). Null when input is zero.
inputNoEchoed input token count.
dev10xNolog₁₀(Leverage). Logarithmic context amplification.
outputNoEchoed output token count.
yield_NoYield (Υ) = (cache_read × output) / input². Headline cascade efficiency. Null when input is zero.
leverageNoLeverage = cache_read / input. Reusable context amplification. Null when input is zero.
velocityNoVelocity = output / input. Output per unit of input. Null when input is zero.
cache_readNoEchoed cache-read token count.
cache_writeNoEchoed cache-write token count.
non_compoundingNoTrue if cache_write is zero (no compounding context).

TDQS

A4.3/5.0
Behavior4/5

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

Annotations already declare readOnlyHint, idempotentHint, and destructiveHint=false. The description adds meaningful extra context: 'without submitting data' and 'No data is persisted,' which go beyond the annotations by clarifying privacy and non-persistence behavior. This is valuable for an agent handling sensitive token-count data.

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

Conciseness5/5

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

The description is three tight sentences with no filler. It front-loads the core action and resource, names outputs in the second sentence, and closes with the key non-persistence guarantee. Every sentence earns its place.

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

Completeness5/5

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

For a simple four-input calculator, the description names every output metric, states the local/non-submitting nature, and notes non-persistence. Since an output schema exists and all parameters are self-describing, nothing essential for correct invocation is missing.

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 the input schema already fully documents input, output, cache_read, and cache_write. The description adds only a restatement that there are four non-negative token counts, which is already visible in the schema, so it does not materially enhance parameter understanding.

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?

The description opens with a specific verb ('Calculate') and resource ('SigRank cascade metrics'), then narrows scope with 'from four non-negative token counts without submitting data.' It lists concrete return values (Yield, Leverage, Velocity, SNR, 10xDEV), making it easy to distinguish from sibling diagnostics.

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

Usage Guidelines4/5

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

The description clearly implies when to use it: when you have four token counts and want local cascade metrics without data submission. It does not explicitly name or exclude sibling tools, but the privacy-preserving 'without submitting data' and 'No data is persisted' context gives an agent enough signal to select it over remote or diagnostic alternatives.

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

All SigRank — AI Operator Benchmarking toolsInspect
Try in Browser
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

B3.4/5.0
Disambiguation2/5

Several tools have significantly overlapping purposes: benchmark_me and compare_to_field both compare a user's cascade to the field, operator_signature and who_operates_like_me both find comparable operators, and rank_paste/rank_windows overlap as metric calculators. Descriptions clarify the output format, but an agent would frequently struggle to pick the right tool.

Naming Consistency3/5

All names are lowercase snake_case, which is readable, but the pattern is mixed: get_leaderboard and simulate_change are verb-first, field_anomaly and operator_gap are noun-first, and rank_if, rank_paste, benchmark_me, and who_operates_like_me break the convention entirely. There is a loose semantic system, but no strong predictable verb_noun pattern.

Tool Count3/5

16 tools sits right at the heavy borderline, and the count feels inflated by overlapping tools that could be consolidated. The domain is broad enough to justify more than a handful of tools, but the duplication makes the set feel heavier than its actual functional surface.

Completeness4/5

The tool set covers the core benchmarking workflow well: reading leaderboard/operator data, computing metrics, comparing to the field, diagnosing weaknesses, simulating changes, and suggesting improvements. There are minor gaps, such as no way to retrieve a user's raw token pillars from a codename for simulation tools that require four pillars, but these are workable.

Resources