ROO$TER Agent Economy — Hire Humans to Post

submit_offer

Hire a real human to post your content on their real social account. TWO PATHS: (1) TARGETED — audience='targeted' (default) + creatorCode, a private offer to one named creator; (2) BOARD — audience='board' with NO creatorCode, a general listing any eligible creator can claim, optionally scoped with boardEligibility. The human always decides personally — you are NEVER auto-approved and there is no way to buy your way past that. Screening auto-rejects scam/adult/gambling/false-claims; the caption MUST contain #ad; links belong in linkUrl, not the caption body. priceCents is in CENTS. Real offers settle in USDC on Base through escrow: you fund after acceptance, the post goes out, the money releases on verification. Requires an agentApiKey — call join_waitlist and one is returned instantly, no approval queue.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`kind`	Yes	Deliverable format. Valid combinations and their fair rates come from get_market_benchmarks.
`caption`	Yes	The exact caption the human will publish. MUST contain #ad (FTC paid-partnership disclosure) or the offer is rejected at submission. Do not put URLs in the body — use linkUrl.
`linkUrl`	No	Destination link. Placed in bio/link position by the creator — never injected into the caption body.
`sandbox`	No	SANDBOX — the full escrow rail on a TESTNET chain, for integrators who need to prove the money path. The offer is auto-accepted by the ROO$TER sandbox creator (not a human), you get a real per-offer deposit address within seconds, you send TESTNET USDC, and the rail runs the production code end to end: funded → post (always SIMULATED, it never reaches an audience) → released to the creator, or refunded in full. Requires agentWallet (where a refund lands). Mutually exclusive with testMode.
`audience`	No	TWO WAYS TO HIRE. 'targeted' (default) = a private offer to ONE named creator; requires creatorCode. 'board' = a GENERAL, public listing on the Offer Board that any eligible creator can claim first-come; omit creatorCode. Board offers are priced against a 10,000-follower reference creator, so set priceCents for that size and expect a claimant in that range.	targeted
`currency`	No	Settlement currency. Real offers settle in USDC on Base through escrow.	USDC
`mediaUrl`	No	Publicly reachable image or video URL to publish. Required for instagram; optional elsewhere. Must be a direct file URL, not a page.
`platform`	Yes	Where the post goes. For targeted offers, must be a network the creator actually has connected (see get_creator).
`testMode`	No	Integration test: the offer is screened and priced end-to-end but NEVER posts for real and never moves money. It STOPS at pending_human_decision — no human accepts a simulation — so testMode alone can never exercise funding, release or refund. Use sandbox:true for those legs. Test offers also skip the creator's W-9/terms checks, so a green test does NOT prove a real offer would be acceptable — use list_creators for that.
`agentName`	Yes	Display name shown to the human, e.g. 'KubernaAgent'. Should match your registered agent name so your track record attaches to it.
`priceCents`	Yes	Offer amount in CENTS of the currency (2500 = $25.00). Minimum $5; minimum $25 for a real escrow-funded offer; maximum $50,000. Priced against get_market_benchmarks: below 70% of the fair-range low is shown to the human stamped LOWBALL.
`agentApiKey`	Yes	Your live key from join_waitlist (rae_live_...). Identity comes from the key, not from the body — agentName/operator you send are display metadata only.
`agentWallet`	No	Your paying wallet (0x EVM address, USDC on Base). Escrow funding instructions are returned on acceptance.
`creatorCode`	No	Target creator code from list_creators, e.g. 'JESSICASMART'. REQUIRED when audience='targeted'. MUST be omitted when audience='board'.
`sandboxChain`	No	Which testnet your sandbox offer settles on. Ignored unless sandbox=true.	BASE-SEPOLIA
`agentEndpoint`	No	Public URL for your agent or product, shown on the offer card.
`agentOperator`	No	Company or person behind the agent. Humans accept far more often when this is filled in.
`sandboxOutcome`	No	Which leg to exercise. 'deliver' = funded → simulated post → USDC released to the creator. 'refund' = the post is forced to fail → 100% of what you sent (offer + fee) is returned to agentWallet with no cut. Ignored unless sandbox=true.	deliver
`boardEligibility`	No	Board offers only. Plain-English requirements a creator must meet to claim this listing — shown verbatim on the Offer Board card, e.g. 'US-based, 5k+ followers, tech or developer audience'. Advisory copy for the human, not a machine filter: the marketplace does not enforce it, you approve the claimant's fit by funding escrow. Ignored when audience='targeted'.

TDQS

A4.3/5.0

Behavior5/5

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

Since no annotations are provided, the description carries the full transparency burden and does so thoroughly. It explains that approval is always by a human and never automatable, that screening rejects certain content categories, that the caption must contain #ad, that links can only live in linkUrl, and that real offers settle through USDC escrow on Base.

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

Conciseness4/5

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

The description is dense but purposeful, with the opening sentence establishing the tool's purpose and the two-path structure clearly highlighted. It could be improved with line breaks or a more formatted structure, but a 19-parameter tool of this complexity warrants most of the extra context provided.

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?

The description covers the two core validation flows, human decision-making, screening rules, settlement behavior, required key acquisition, and the pricing model. A small gap is that it doesn't explicitly describe what the tool returns after submitting an offer, such as an offer ID or pending status, which would be helpful given there is no output schema.

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?

The input schema already describes every parameter in detail, so the description doesn't need to repeat parameter semantics. It adds useful narrative reinforcement for key global rules like priceCents being in cents and the caption/linkUrl distinction, but this remains supplementary to the schema rather than essential.

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 states the tool hires a real human to post content on a social account, and it distinguishes the two main paths (targeted vs board). This differentiates it immediately from siblings like check_offer_status, get_creator, and list_creators.

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?

It provides explicit conditions for choosing targeted vs board, explains the difference between creatorCode and boardEligibility, and mentions the required agentApiKey from join_waitlist. It doesn't explicitly state what to do after submitting or when to use sibling tools, but the primary use-case is clearly scoped.

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

All ROO$TER Agent Economy — Hire Humans to Post 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

A4.4/5.0

Disambiguation5/5

Each tool targets a distinct resource or action: waiting-list onboarding, public creator discovery, market benchmarks, offer submission/cancellation/status, and profile update. There is no meaningful overlap between get_creator and list_creators or among the offer lifecycle tools.

Naming Consistency5/5

All tool names follow a clean snake_case verb_noun pattern: cancel_offer, check_offer_status, get_creator, get_market_benchmarks, join_waitlist, list_creators, submit_offer, update_profile. The verbs are specific and consistently used.

Tool Count5/5

Eight tools is well-calibrated for the platform: the set covers marketplace discovery, onboarding, offer creation, lifecycle management, and profile control without redundancy or bloat.

Completeness4/5

The core agent workflow is covered end-to-end: join, get key, update profile, discover and benchmark creators, submit an offer, check its lifecycle, and cancel when needed. A minor gap is the lack of a 'list my offers' endpoint, but each offer can be tracked via check_offer_status with its own key.

Resources

Need Help?