Skip to main content
Glama

Robyn — Gasless Cross-Chain for AI Agents

by io.github.ox31461

NULL forward secrecy — fetch a peer prekey bundle (consumes one one-time key)

null_prekeys_fetch

Fetches {spk, sigSpk, otk, otksLeft} for a recipient so you can start an X3DH-PQ forward-secret session (SDK: fs.startOutbound). ALWAYS fetch by META-ADDRESS: the sigSpk check verifies against the meta-address spendPub (sha256("null-fs-spk/v1:"+spk), secp256k1) and CANNOT run on a handle fetch (the SDK tags those fsUnverified — treat as unauthenticated keys). The nonce is MANDATORY and idempotent: a retry with the SAME nonce returns the SAME one-time key instead of draining the pool; use a fresh random 12-64 url-safe chars per new session. If otk comes back null the pool is empty or rate-limited (issuance is capped per recipient to slow drain attacks): proceed SPK-only — still forward-secret via the ratchet, slightly weaker for the first message — and the recipient replenishes automatically. A failed sigSpk check means a key-substituting relay: abort to the static seal and treat it as an incident, not a fallback. Read (consumes one otk per fresh nonce).

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
nonceYes12-64 url-safe chars; SAME nonce on retry = same key returned, no drain
recipientYesst:eth:0x… meta-address (REQUIRED for the SPK signature check; a handle fetch returns unverifiable keys)

Schema Changelog

Changes observed during successful MCP inspections. Dates show when Glama detected each change.

  1. Changed1 schema field changed
    • changedInput schema / properties / recipient / description
      Previous value: -"st:eth:0x… meta-address (preferred — enables the SPK signature check) or a handle"New value: +"st:eth:0x… meta-address (REQUIRED for the SPK signature check; a handle fetch returns unverifiable keys)"
  2. Added

TDQS

A4.9/5.0
Behavior5/5

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

With no annotations provided, the description carries the full burden and succeeds: it discloses one-time-key consumption, nonce idempotency, per-recipient rate capping, the null-otk fallback, and that a failed sigSpk check signals a key-substituting relay incident. No major side effect or safety-relevant trait is hidden.

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 core purpose and the most urgent requirement — fetch by meta-address, not handle — are front-loaded. Each subsequent sentence covers a distinct operational or security concern, and there is no filler despite the description's length.

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 security-sensitive fetch with no output schema, the description names the return tuple, defines otk-null semantics, covers rate limiting, idempotency, and failure handling. An agent has enough context to call the tool correctly and handle every documented outcome.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, but the description adds unique meaning: a fresh nonce is required per new session, the sigSpk verification path is spelled out, and pool-depletion behavior is explained. There is some duplication of the schema's own parameter descriptions, which keeps this from a perfect score.

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 concrete action and resource: 'Fetches {spk, sigSpk, otk, otksLeft} for a recipient' to start an X3DH-PQ forward-secret session. It clearly distinguishes this from peer tools such as null_prekeys_publish by focusing on fetching a peer's prekey bundle for outbound sessions.

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

Usage Guidelines5/5

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

It explicitly states when to use the tool (starting a forward-secret session), mandates meta-address fetches over handle fetches, and gives actionable edge-case guidance: proceed SPK-only if otk is null, abort to the static seal if sigSpk verification fails. The agent does not have to infer when this tool is appropriate.

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

All Robyn — Gasless Cross-Chain for AI Agents 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

A3.6/5.0
Disambiguation2/5

Many tools have overlapping purposes, such as robyn_quote, robyn_prepare, robyn_settle, robyn_cross_chain, robyn_plan, robyn_preflight, and robyn_preflight_full, all involving routing or quoting. Although descriptions are detailed, the high number of subtly different tools makes selection error-prone.

Naming Consistency5/5

All tools follow a consistent <prefix>_<verb/noun> pattern, with robyn_ and null_ as clear category prefixes. Verbs like prepare, submit, verify and nouns like quote, plan, mesh are highly predictable throughout.

Tool Count2/5

With 70 tools, the count far exceeds the 25+ heavy threshold. While the service scope is broad (payments, messaging, netting, Plus, yield, etc.), many tools could be merged or tiered, making the set feel overly large and burdensome for an agent.

Completeness4/5

The surface covers the full lifecycle from quoting, planning, submitting, verifying, auditing to messaging, custody, yield, and mandates, plus sandbox, lint, and recipes. A few advanced operations like cancellation or revocation are missing, but no critical gaps remain.

Resources