MCPFax Agent Continuity

Attempt count and backoff that survives a crash

retry_state

Idempotent

Durable retry bookkeeping, so backoff survives the session that died. Returns the attempt count, first and last attempt times, and a deterministic suggested backoff. Pass record:true to count this attempt. $0.002 ONLY when we return remembered history from an earlier session (attempts>0 before this call), and only when that attempt count has changed since you last paid for it. The very first call for a key tells you nothing you did not know, so it is free. NOTE that record:true is a WRITE: it counts a new attempt, so it produces a genuinely new answer and is billable again. Use the default record:false to read the history without changing it. Costs $0.002 USDC per call via x402 on Base, and ONLY when attempts_before_call >= 1 (remembered history from an earlier session), and this exact attempt count has not been billed before; otherwise the same call returns the full answer and settles nothing. Authenticate with Authorization: Bearer , or pass agent_key as an argument if your host cannot set headers. Equivalent HTTP route: POST /v1/retry-state.

Input Schema

TableJSON Schema

Name	Required	Description
`key`	Yes	What is being retried. Namespace-scoped, hashed before storage. Example: 'sync-vendor-7'.
`scope`	No	Optional unit of work, so resume_packet can report what is being retried. Example: 'permit-review-2026-08'.
`record`	No	Count this attempt before answering. Default false. Example: 'true'.
`agent_key`	No	Your agent_secret, if your MCP host cannot set the Authorization header. Prefer the header.
`ttl_seconds`	No	Retention. Default 604800 (7d), max 2592000. Example: '604800'.

TDQS

B3.3/5.0

Behavior1/5

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

Annotation Contradiction: The description states that record:true is a WRITE that counts a new attempt and produces a genuinely new answer, implying repeated calls with the same payload have different effects. This contradicts the annotation idempotentHint=true, which implies repeated identical calls are idempotent. Per the rubrics, any description that contradicts annotations receives a score of 1.

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

Conciseness2/5

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

The description is overly long and repetitive, especially around billing. The $0.002 cost and when it applies is restated multiple times in different forms. While the core intent is front-loaded, the redundancies undermine conciseness and structure.

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

Completeness4/5

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

With no output schema, the description lists return values (attempt count, first/last times, suggested backoff) and explains authentication, billing, and the semantic distinction of record. It omits edge-case behavior like missing keys, but overall provides a robust operational picture for an agent.

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?

Input schema coverage is 100%, so parameters are already well documented. The description adds billing semantics and clarifies record behavior, but mostly reiterates what the schema says (e.g., key hashing, record default false). It doesn't materially surpass the schema, so baseline 3 is appropriate.

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 clearly identifies 'retry_state' as durable retry bookkeeping that survives crashes, and enumerates exactly what it returns (attempt count, first/last times, suggested backoff). It is distinct from sibling tools like work_fail or should_continue, which address different workflow concerns.

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?

Provides explicit guidance: use record:true to count an attempt, record:false to read history without mutation. Also explains authentication options. It doesn't reference sibling tools as alternatives, but the usage context is clear enough for an agent to decide when to call this tool.

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

All MCPFax Agent Continuity tools Inspect

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

A4/5.0

Disambiguation5/5

Every tool targets a clearly distinct mechanism—barriers, budgets, checkpoints, claims, heartbeats, inboxes, leases, work queues, etc. Even closely related pairs like claim/complete, work_done/inbox_ack, and barrier_status/work_status are explicitly differentiated in their descriptions, so an agent can reliably select the right tool.

Naming Consistency4/5

The dominant pattern is object_verb (e.g., checkpoint_get, work_push, seen_add, lease_release), which is consistent and readable. A few single-word verbs (claim, complete, send) and noun-only names (resume_packet, retry_state, whoami) break the pattern, but they are few and still intuitive.

Tool Count2/5

With 33 tools, the surface is well beyond the 25-tool threshold for 'too many' and will feel heavy for agents to explore, even though the tools are organized into subdomains. The scope is broad, but the sheer number makes the server less approachable and increases the chance of misselection.

Completeness5/5

The tool set provides thorough lifecycle coverage across its domains: checkpoints get/put, work queue push/take/done/fail/status, inbox poll/ack/nack/schedule, lease acquire/renew/release, barrier create/signal/status, dedup add/check, watermark get/set, retry state, budget check/record, and idempotency claim/complete. There are no obvious dead ends or missing essential operations.

Resources

Need Help?