Skip to main content
Glama
Sign In

Wever Labs Agent Products

by io.github.CodeWever
OverviewDiscussionsAdmin

Rollback path check

wever_rollback-path-check
Read-only

Agent Recovery: submit the effects a task can leave outside itself (a payment, a message, a booking, an order, a record, an access change or another external write), each with how it is undone (automatically, by a runbook, by a compensating action, not at all, or it cannot be undone), the reversal as written, how long it stays possible, whether it is marked irreversible before the task runs, whether it waits for a person’s approval, whether the reversal has been run and when it was written down, plus when a task first failed and needed its effects undone. The signed report says for each effect whether it can be reversed, whether the reversal was written before the first failure and which effects are marked irreversible up front, and names every effect with no reversal and no mark, every irreversible effect left unmarked and every reversal never written down, then every irreversible effect run without approval, every reversal never run and every reversal written after the first failure, with the fix for each. This reads the way back for a failed task’s effects: money back from a provider for a failed paid call is wever_refund-path-check, backups and restore steps for data are wever_recovery-readiness-check and what a stop does to in-flight actions is wever_kill-switch-check. Verdict undo_undefined, rollback_gaps, rollback_unverified or undo_defined. POST /api/rollback-path-check. Reads only the material in the call: no network request, fetch, money movement, message, sharing, configuration change or external write. Paid URL: 0.10 USDC per call on Base via x402 exact, with the shared 10 signed-wallet free calls per rolling 30 days. Bind X-Wever-Wallet-Proof to the exact JSON arguments and the tool route. Returns a report with an Ed25519 signed receipt. Contract: /api/agent-recovery-contract. Human page: https://weverlabs.com/agent-recovery/.

Input Schema

NameRequiredDescriptionDefault
labelNoOptional name recorded in the report.
systemNoThe task system, such as the booking pipeline.
effectsYesEach effect a task can leave outside itself.
first_failure_atNoWhen a task first failed and needed its effects undone.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Added

TDQS

A3.6/5.0
Behavior4/5

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

Annotations already declare readOnlyHint=true, openWorldHint=false, destructiveHint=false, and idempotentHint=false, so the safety profile is covered. The description adds substantial context beyond those: the exact HTTP route, the paid pricing model (0.10 USDC on Base via x402 exact with 10 free signed-wallet calls per 30 days), the X-Wever-Wallet-Proof binding requirement, the Ed25519 signed receipt, and the explicit no-network/no-money-movement guarantee. That is meaningful operational detail an agent cannot get from the annotations.

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 a single sprawling run-on sentence that enumerates every effect field twice (once as an input list, once as report output) before the routing and pricing details. The verdict values and sibling routing are buried in the middle, and the 'POST /api/...' plus payment proof instructions trail after the substantive content. It is over-long and poorly front-loaded for what it conveys.

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?

There is no output schema, so the description carries the return-value burden, and it does name the report contents, verdict values (undo_defined, rollback_gaps, rollback_unverified, undo_undefined), and the signed receipt. Combined with the contract URL and human page, an agent knows what to expect back. The only gap is that the exact report field structure is described prose-wise rather than precisely, but coverage is adequate.

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 schema fully documents label, system, effects, first_failure_at, and every nested effect field including the reversal enum. The description's long enumeration of effect attributes largely restates what the schema already contains (how it is undone, the reversal as written, how long it stays possible, approval, marked irreversible, tested, defined_at). It adds little syntax or semantics beyond the schema, so the baseline 3 applies.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description names a specific analysis verb (submit effects and get back a signed report on reversibility, written-before-first-failure ordering, and unmarked irreversibility) over a specific resource (a task's external effects). It explicitly distinguishes itself from three named siblings (refund-path-check for money, recovery-readiness-check for backups, kill-switch-check for in-flight actions), so an agent can differentiate it. It falls short of 5 only because the opening 'Agent Recovery: submit the effects...' spends several clauses enumerating input fields before stating what the tool computes.

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 gives a clear usage context: pass in a failed task's effects to determine whether the undo path is defined, verified, and documented before first failure. It names the alternative tools and what each covers ('money back from a provider... is wever_refund-path-check', backups are recovery-readiness-check, stop behavior is kill-switch-check), which is genuine routing guidance. It does not state explicit when-not-to-use conditions or prerequisites beyond that.

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

All Wever Labs Agent Products toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources