Skip to content
verify mcp Beta VerifyMCP is currently in beta. If you notice any issues, get in touch and we’ll put it right.

RentAHuman

NPM · RENTAHUMAN-MCP

Hire humans for digital microtasks, research, QA, data evaluation, and physical-world tasks.

Available components

−8 this week 67 Trust /100

Last scanned (15 hours ago)

Scores higher than 38% of npm servers

Category Workflow Automation

Trust breakdown (7 categories)

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →

Supply Chain Security90
  • No malware found by supply-chain analysis.Pass
  • CVE check failed: a known high-severity CVE affects toml 4.3.0, reached via @effect/cli > toml. A fixed version is available. View diagnostics → Fail
  • No install/post-install scripts declared.Pass
  • 35 of 148 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency19
  • Repository check failed: the declared repository URL returned HTTP 404. See how to fix → View diagnostics → Fail
  • Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
  • Clear OSI-approved license (MIT).Pass
  • Actively maintained (last published 3 days ago).Pass
  • Security-disclosure policy not yet verified: we couldn't inspect the source repository.Unverified
Schema Quality & AI Usability77
  • 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).Pass
  • AI-judged instruction clarity (excellent).Pass
  • Context-footprint check failed: tool/resource definitions use about 20535 tokens (~285/item across 72 items; 69 tools + 3 resources), over budget; trim descriptions and params. See how to fix → Fail
  • Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management36
  • Stability check failed: the tool surface changed between 3.14.0 and 5.0.0: 59 tool removals, 11 breaking changes, 28 additions. See how to fix → Fail
Tool Coverage96
  • 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
  • 87% of tool parameters carry a description.Partial
  • Structured output schemas are declared (96% of tools); any adoption earns full credit.Pass
Tool Safety100
  • No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
  • All 8 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.Pass
  • An AI judge read all 71 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
  • Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
What we tested

Checks we ran (15)

  • Malware scan
  • Known CVEs
  • Install scripts
  • Dependency health
  • Source repository
  • Provenance
  • License
  • Maintenance
  • Schema quality
  • Stability
  • Tool coverage
  • Injection markers
  • Destructive annotations
  • Judged manipulation
  • MCP protocol

Checks we could not complete (1)

These are gaps on our side, not findings about this package. We only credit what we can confirm, so a category made up entirely of them scores 0.

  • Security disclosure in Provenance & TransparencySecurity-disclosure policy not yet verified: we couldn't inspect the source repository.

Does not apply to this package (2)

  • Endpoint SecurityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.
  • Transport & ReachabilityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.

The score does not measure everything. What the score does not measure →

Changelog

Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.

  • 18 Nov 26 −8
    • Schema quality: 253 → 285 ▼ functional
    • First check of Tool coverage: 96 functional
    • Destructive annotations: All 8 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation. functional
    • Package version: 4.10.0 → 5.0.0 functional
  • 13 Nov 26 +1
    • Stability: fail → unverified ▼ security
    • Tool safety: pass → unverified ▼ security
    • Capabilities: pass → unverified ▼ functional
    • Tool coverage: 100 → unverified ▼ functional
    • Schema quality: 100 → unverified ▼ functional
    • Package version: 4.9.0 → 4.10.0 functional
  • 12 Nov 26 +15
    • Malware scan: unverified → pass ▲ security
    • Package version: 4.8.0 → 4.9.0 functional
  • 11 Nov 26 −15
    • Malware scan: pass → unverified ▼ security
    • CVE-2023-54404 no longer affects this package ▲ security
    • Package version: 4.7.0 → 4.8.0 functional
  • 9 Nov 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 80 to 83.

  • 7 Nov 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 73 to 76.

  • 6 Nov 26 −3

    No change was recorded against any check on this day. Stability & Change Management went from 92 to 73.

  • 5 Nov 26 0
    • Stability: fail → unverified ▼ security
    • Tool safety: pass → unverified ▼ security
    • Capabilities: pass → unverified ▼ functional
    • Tool coverage: 100 → unverified ▼ functional
    • Schema quality: 100 → unverified ▼ functional
    • Package version: 4.6.0 → 4.7.0 functional
Diagnostics

Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.

Captured 21 Nov 2026 · Analysed npm/rentahuman-mcp@5.0.0

Provenance No attestation

The registry publishes no build provenance for this version, so there is nothing to verify.

Result No attestation
Ecosystem npm

Background: How many MCP packages publish verified provenance →

Vulnerabilities 4 findings
ID CVE Severity Vector Fix available
GHSA-4qx8-w7c3-pj3p CVE-2026-114141 high yes
GHSA-9j65-hcg3-gc64 CVE-2026-114143 high yes
GHSA-9qmh-m7c6-f92p CVE-2026-114144 high CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H yes
GHSA-h895-cgqg-jf52 CVE-2026-114142 high yes

Background: What a vulnerability scan can and cannot prove →

Dependencies 148 packages
Packages resolved 148
Stale 35
Tree resolution Complete

Background: SBOMs and build attestations, explained →

MCP tools · 69 exposed · ~20,252 tokens

The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →

Tool Tokens
accept_application ~383

Accept one pending application and make a hire; the worker is paid later. An already-funded spot accepts hosted sign-in; paying for a spot requires an API key. If the applicant proposed another price, accepting charges that price without confirmation. Use for: hiring one applicant of your bounty. Not for: paying a worker (pay_hire); ending a hire (cancel_hire). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: accept_application({"applicationId":"a3Fh8kQz1WpL0mNc"}) Repeat: Once this has worked, a repeat does not do it again while that result still stands: it is refused, or answers with what is already there. Errors: spots_full, wallet_short, payment_page_open, held_pay_not_shown, payment_processing, charge_mismatch, payment_needs_attention, spot_not_paid_for, application_not_pending, application_changed, bounty_not_open, accept_in_progress, check_not_passed, applicant_unavailable, account_frozen, spending_limit_reached, account_setup_required; each error carries `fix`: what to do next.

NameTypeReqDescription
applicationIdstringyesThe application id.
idempotencyKeystring–A unique string for this call. Repeat it when you retry, so the call runs once.
messagestring–A note the applicant reads with the acceptance. Markup, script links and outer spaces are removed before it is stored, so the applicant may read fewer characters than you sent. Only when money is alr…
NameTypeReqDescription
actionsarray–Tools you may call on this hire now. Each tool still checks.
bountyobjectyes–
conversationIdstring––
costobject–What this hire costs you as its held money stands now: pay plus service fee. Left out when nothing is held: the fee of a pay-later hire is set when you pay.
dueobject––
hireIdstringyesThe hire that exists for this application now; read it with get_hire. It is not a receipt that this call made it: an earlier call, or a payment arriving, may have.
payStatusstringyes–
statusstringyesoffered: waiting for the worker to confirm. active: working, or fixing a redo. submitted: work waits for your review, or the work is done and its pay waits for you to release or pay it. paid: the pay…

No examples provided.

block_human ~291

Stop one person from applying to your bounties. Blocking also takes them off your preferred list. Blocking again only replaces the reason. Use for: keeping someone you do not want to work with away from your bounties. Not for: letting them apply again (unblock_human); seeing who you blocked (list_blocked_humans). Needs: an API key. Hosted sign-in cannot call this yet. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: block_human({"humanId":"h2Lw9QxT4mZp7Rk","reason":"Missed two shifts"}) Repeat: A repeat with the same input, and the same key if you sent one, does not act twice: it returns the same result, finishes the first call, or is refused. Errors: not_found, list_full, account_setup_required; each error carries `fix`: what to do next.

NameTypeReqDescription
humanIdstringyesThe human id, as `id` in list_humans.
reasonstring–Why you block them, for your own list. They never see it. Up to 280 characters. Left out: none, and a repeat without one clears the earlier one.
NameTypeReqDescription
blockedAtstringyesISO 8601 time with an offset, for example 2026-11-20T09:00:00-08:00.
humanIdstringyes–
reasonstring–Why you blocked them. Missing when there is none.

No examples provided.

book_sign_campaign ~466

Book people to hold your sign on San Francisco (or New York) street corners through the RentAHuman Signs API. Use this instead of create_bounty for any sign-holding, sign-spinning or human-billboard job: you give the design, we print it and handle pickup. Applicants are screened by AI, hired sign holders get a shift briefing and check-ins, and each holder is paid once RentAHuman ops or you approve their work. Requires RENTAHUMAN_API_KEY. Always call with dryRun omitted or true first: it validates and prices the campaign and checks capacity without booking or charging, and returns earliestStart (shift dates must be at least the minimum lead time ahead). Show the user the price and dates. Then call again with dryRun: false, confirm: true, an operationId and the artwork (url or base64; PNG, JPEG or PDF): it uploads the design, books the campaign and pays from the wallet first; if the wallet is short it returns a hosted checkout URL for a human to pay. Card details are never taken. Reuse the operationId to retry. Track the campaign at statusPath (GET).

NameTypeReqDescription
artworkobject–The sign design. Required to book; optional for a quote.
citystring––
confirmboolean–Required true for a live booking, which pays from the wallet first.
datesarrayyesan array of at least 1 item(s)
dryRunboolean–Defaults to true: quote only. Pass false (with confirm: true) to book and pay.
endHourintegeryesan integer
laborCentsPerHourinteger–an integer
operationIdstring–Idempotency key and campaign ID (16-64 letters, digits, _ or -). Reuse it to retry a booking safely; a booked campaign is returned instead of booked twice.
signReturnobject–Where the crew leaves the printed signs after the last day, and what to do.
specialInstructionsstring––
spotsarrayyesOne public-sidewalk point per sign holder, inside the city (max 20).
startHourintegeryesan integer

No output schema declared.

No examples provided.

cancel_bounty ~581

End a bounty and receive its cancellation quote: what is quoted back and what each hired worker keeps. Use for: ending a bounty. Use preview: true for a quote. If confirmation_required, call again with only bountyId, confirmationToken and the same idempotencyKey within 5 minutes. If processing, retry with the same key. Check returned money with get_wallet and list_wallet_transactions. Every call counts toward the daily limit, shared with edits. Not for: ending one hire (cancel_hire); work that was not done (open_dispute). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: cancel_bounty({"bountyId":"b7Kq2mXw9TzR4nLp","reason":"not_enough_applicants","idempotencyKey":"5f0c7a2e-8d4b-4e1a-9c3f-2b6d1e8a7f40"}) Key: the idempotencyKey in the example is a sample. Replace it with a new unique key for each new call, and keep that key for the call’s retries and any confirmation it asks for. Repeat: A repeat with the same input, and the same key if you sent one, does not act twice: it returns the same result, finishes the first call, or is refused. Errors: not_cancellable, confirmation_expired, refund_changed, cancellation_in_progress, idempotency_key_reused; each error carries `fix`: what to do next.

NameTypeReqDescription
bountyIdstringyesThe bounty id, or a pasted rentahuman.ai/bounties/... link.
confirmationTokenstring–The token of a confirmation_required reply. Send it with bountyId only, and an idempotencyKey as on every call, within 5 minutes, to carry out that cancellation. The token decides which cancellation…
detailsstring–Your own words, needed when reason is other and refused with any other reason. Up to 500 characters.
idempotencyKeystringyesA unique string for this call. Repeat it when you retry, so the call runs once.
previewboolean–true returns the quote and a confirmationToken and never starts a cancellation. A cancellation that was already confirmed and is still being finished may be finished by any call, this one included; t…
reasonstring–Why you are ending the bounty. Needed on every call without a confirmationToken: there is no default.
NameTypeReqDescription
bountyobjectyes–
confirmationTokenstring––
expiresAtstring–ISO 8601 time with an offset, for example 2026-11-20T09:00:00-08:00.
quoteobjectyes–
statusstringyesconfirmation_required: nothing was cancelled; call again with only bountyId and confirmationToken, and the same idempotencyKey, which is still required. cancelled: done; book the quote once. A repeat…

No examples provided.

cancel_hire ~451

End one hire. The rest of the bounty keeps running. You get back what the cancellation rules allow. Use for: ending a hire before work is submitted. If confirmation_required, call again within 5 minutes with hireId, confirmationToken from cancellation.confirmation.token, and the same idempotencyKey. If processing, retry with the same key. Read cancellation.refund for the refund quote, or cancellation.held for where the held money went. Not for: ending the whole bounty (cancel_bounty); work that was not done (open_dispute). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: cancel_hire({"hireId":"a3Fh8kQz1WpL0mNc","idempotencyKey":"5f0c7a2e-8d4b-4e1a-9c3f-2b6d1e8a7f40"}) Key: the idempotencyKey in the example is a sample. Replace it with a new unique key for each new call, and keep that key for the call’s retries and any confirmation it asks for. Repeat: A repeat with the same input, and the same key if you sent one, does not act twice: it returns the same result, finishes the first call, or is refused. Errors: not_cancellable, only_spot, submission_waiting, confirmation_expired, refund_changed, cancellation_in_progress, idempotency_key_reused; each error carries `fix`: what to do next.

NameTypeReqDescription
confirmationTokenstring–The token of a confirmation_required reply. Send it with hireId and the same idempotencyKey within 5 minutes to carry out that cancellation. Default: none, which asks to end the hire.
hireIdstringyesThe hire id, from list_hires or accept_application.
idempotencyKeystringyesA unique string for this call. Repeat it when you retry, so the call runs once.
NameTypeReqDescription
actionsarray–Tools you may call on this hire now. Each tool still checks. Left out when there are none, and when status is left out: then read the hire with get_hire.
cancellationobjectyes–
endedReasonstring–Why the hire ended, once it has. Left out while it has not ended, and when status is left out because the hire could not be read after this call.
hireIdstringyes–
payStatusstring–Left out only when the hire could not be read after this call: then status, payStatus and actions are all left out. Read it with get_hire.
statusstring–Left out only when the hire could not be read after this call: then status, payStatus and actions are all left out. Read it with get_hire.

No examples provided.

create_api_key ~225

Make a new API key for your account. The secret is shown once, in this reply. Its id is in list_api_keys. Use for: giving another agent or machine its own key, so you can revoke it alone later. At most 25 active keys. Not for: finding the id of a key (list_api_keys); turning a key off (revoke_api_key). Needs: an API key. Hosted sign-in cannot call this yet. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: create_api_key({"name":"Booking agent"}) Repeat: A repeat may act again on the thing as it stands then. Read it first. Errors: key_limit_reached, account_setup_required; each error carries `fix`: what to do next.

NameTypeReqDescription
namestringyesA name to tell this key apart, 1 to 50 characters, not only spaces. Spaces around it are dropped.
NameTypeReqDescription
keyobjectyes–
secretstringyesThe new key. Shown in this reply only: store it now. Send it in `X-API-Key`.

No examples provided.

create_bounty ~731

Post a bounty. It is paid for from your wallet at once when the balance covers it; use dryRun to see the cost first. Use for: posting new work. A short wallet saves a draft with `payUrl`, when a pay page could be made. Without `payUrl`, add money with fund_wallet, then call fund_bounty. Send an idempotencyKey to reuse the first bounty and ignore new input. Unkeyed retries may post again or be refused as duplicates; check list_bounties first. Not for: paying for a draft (fund_bounty). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: create_bounty({"title":"Photograph the storefront of 5 coffee shops","description":"Walk to each of the 5 listed coffee shops and take one clear photo of the storefront in daylight.","category":"creative-gigs","pay":{"type":"fixed","amount":1999,"estimatedHours":2},"spots":{"total":3},"location":{"country":"US","isRemoteAllowed":true},"proof":{"criteria":"Five photos, one per shop, each showing the shop sign.","types":["photo"]}}) Repeat: A repeat may act again on the thing as it stands then. Read it first. Errors: listing_refused, pay_below_minimum, duplicate_listing, spending_limit_reached, account_setup_required, wallet_short, create_in_progress, photos_not_available; each error carries `fix`: what to do next.

NameTypeReqDescription
aiobject––
applicationsobject––
bonusesarray––
categorystringyesThe kind of work, as in the website’s category picker: workers browse bounties by it. Required, with no default: pick the closest. computer-gigs is work done at a computer; photographing a place is c…
checksobject––
countriesobject––
descriptionstringyesa string at most 5000 character(s) long
dryRunboolean–true returns the preview and the cost and posts nothing. A dry run counts against your posting limits, and passing it does not promise the real post passes: some checks run only on a real post. Defau…
idempotencyKeystring–A unique string for this call. Repeat it when you retry, so the call runs once.
imageUrlsarray––
locationobject–Omit this group to post a remote job (work from anywhere).
messagesobject––
payobjectyes–
postAsOrganizationboolean–true posts on behalf of your organization; you still pay from your own wallet. Refused when you belong to no organization. Default false.
proofobjectyes–
requirementsarray––
scheduleobject––
signagestring–Sign-holding jobs only: send "poster_provided" when you bring your own signs. Default: not sent.
skillsarray––
spotsobject––
startsAtstring––
timeZonestring–IANA time zone for startsAt and every shift of schedule.
timersobject––
titlestringyesa string at most 200 character(s) long
NameTypeReqDescription
actionsarray–Tools you may call on this bounty now. Each tool still checks.
bountyIdstring––
chargeobject–What this post took from your wallet. Left out when it took nothing, or when isChargeUnread is true.
dryRunobject–Set on a dry run only: the bounty a real post would make, and what it would cost. Nothing is posted then, and bountyId, status, charge and payUrl are left out.
isChargeUnreadboolean–true when the post may have taken money but the amount could not be read: check list_wallet_transactions. Left out otherwise.
payUrlstring–Set when the bounty is a draft because your wallet did not cover it: a page where it can be paid for. It can be left out then too; fund_bounty pays for the draft from your wallet once it covers the t…
statusstring––

No examples provided.

create_humanization ~599

Order one human to rewrite a text by hand, without AI. It is paid from your wallet at once when the balance covers it. Use for: ordering a hand-written rewrite of private text. When your wallet does not cover `cost.total`, the order waits as awaiting_payment and fund_bounty pays it later. The same idempotencyKey returns the first order, ignores new input, and never orders or charges twice. Not for: paying for an order awaiting payment (fund_bounty); reading where an order stands (get_bounty). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: create_humanization({"sourceText":"Our quarterly results exceeded expectations across every region, driven by strong demand.","transformation":"general_rewrite","subject":"quarterly update","turnaroundMinutes":120,"price":{"amount":2500,"currency":"usd"},"idempotencyKey":"5f0c7a2e-8d4b-4e1a-9c3f-2b6d1e8a7f40"}) Key: the idempotencyKey in the example is a sample. Replace it with a new unique key for each new call, and keep that key for the call’s retries and any confirmation it asks for. Repeat: A repeat with the same input, and the same key if you sent one, does not act twice: it returns the same result, finishes the first call, or is refused. Errors: humanization_unavailable, listing_refused, pay_below_minimum, duplicate_listing, spending_limit_reached, account_restricted, account_setup_required, wallet_short, create_in_progress; each error carries `fix`: what to do next.

NameTypeReqDescription
applicantScreeningobject–Applicants first rewrite a sample of this AI-written text by hand, with a screen recording. The sample is public on the listing.
idempotencyKeystringyesA unique string for this call. Repeat it when you retry, so the call runs once.
instructionsstring–Private notes for the writer.
priceobjectyesWhat the writer is paid, from $1 to $1,000,000. Our fee is added on top.
requireResumeboolean–true asks each applicant for a resume. Default false.
sourceTextstringyesPrivate text for the accepted writer; 10,000 characters maximum including instructions and the two added newlines when instructions is supplied.
subjectstring–A short PUBLIC topic shown in the listing title, such as "essay on AI infrastructure". Never private content.
transformationstringyes–
turnaroundMinutesintegeryesEstimated writing time, not a deadline from ordering.
NameTypeReqDescription
chargeobject–What the first call with this idempotencyKey took from your wallet. Left out when it took nothing, and on a repeat 24 hours or more later: see list_wallet_transactions.
costobjectyes–
humanizationobjectyes–
payUrlstring–Set when the first call with this key left the order awaiting_payment: a page where the shortfall can be paid. It can be left out then too; fund_bounty pays for it from your wallet.

No examples provided.

create_qa_run_template ~769

Starts runs that spend money from your wallet: people test your website for pay, the first run within about ten minutes, later runs on your cadence with no further call. Use for: paid human testing of a website. Each run takes costPerRun from your wallet; up to periodCap a month. With an idempotencyKey, within 24 hours the same key returns the first template and ignores new input. After 24 hours or without a key, a repeat saves a second template; both run and charge. Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: create_qa_run_template({"name":"Checkout smoke test","targetUrl":"https://shop.example.com/checkout","instructions":"Buy the cheapest plan with the test card and report every error you hit, with steps.","cadence":"weekly","testerCount":2,"payPerTester":{"amount":2000,"currency":"usd"},"budgetPerRun":{"amount":4720,"currency":"usd"},"periodCap":{"amount":23600,"currency":"usd"},"submissionMode":"video"}) Repeat: A repeat may act again on the thing as it stands then. Read it first. Errors: account_setup_required, create_in_progress; each error carries `fix`: what to do next.

NameTypeReqDescription
allowedCountriesarray–Two-letter country codes testers may be in. Default: any.
applicantScreeningobject–What an applicant must give: a live camera video, a LinkedIn profile, a PDF, DOCX or TXT resume. Each defaults to false.
autoModeboolean–Default true: we hire and review every tester, remove poor work and pay good work. false asks you for every hiring and pay decision.
budgetPerRunobjectyesA ceiling per run, $1 to $5,000. It must cover payPerTester plus the 18% fee (rounded to the cent) times testerCount. A run takes costPerRun, not this.
cadencestringyesHow often a run starts. once runs one time; the others repeat with no further call.
cadenceDays––Days between runs, 1 to 30. Required for custom, refused for any other cadence.
checklistarray–Steps every tester must cover; each becomes a point their work is graded on.
idempotencyKeystring–A unique string for this call. Repeat it when you retry, so the call runs once.
instructionMediaarray–Screenshots or short clips shown with the instructions.
instructionsstringyesWhat every tester does.
namestringyesa string at most 120 character(s) long
payPerTesterobjectyesWhat each tester gets, $1 to $5,000. Our 18% fee is added on top.
periodCapobjectyesMonthly spending ceiling, at least budgetPerRun; equal to it for once. Recurring runs need monthly coverage, capped at $50,000; an error gives the minimum.
requiredCredentialsarray–Skills or experience an applicant must show.
submissionModestringyesWhat each tester hands in.
targetUrlstringyesThe website to test, http or https. It is checked when saved: when it cannot be reached, the template is saved paused.
testerCountintegeryesa number between 1 and 20
testerStartMessagestring–Sent privately to each tester once hired, for example test sign-in details.
NameTypeReqDescription
costPerRunobjectyesWhat each run takes from your wallet when it starts: testerCount times (payPerTester plus the 18% fee, rounded to the cent).
templateobjectyes–
walletIsShortByobject–How much your wallet lacked for costPerRun when the template was saved. A run your wallet cannot cover waits for funds until the next run is due. Left out when the wallet covered it.

No examples provided.

create_taste_run ~711

Ask a number of people to choose between 2 to 6 options, and pay each one. The cost is respondentCount times (pay plus 18% of pay, rounded), taken from your wallet in this call. Use for: a vote or taste test. If funds are short, the saved run starts automatically after funding. Send an idempotencyKey to reuse the first run and ignore new input. Unkeyed retries may charge again. Not for: adding money to your wallet (fund_wallet); finding a run you started earlier (list_taste_runs). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: create_taste_run({"question":"Which logo feels more trustworthy, and why?","options":[{"label":"Blue","url":"https://example.com/logo-blue.png"},{"label":"Green","url":"https://example.com/logo-green.png"}],"respondentCount":10,"payPerRespondent":{"amount":100,"currency":"usd"}}) Repeat: A repeat may act again on the thing as it stands then. Read it first. Errors: spending_limit_reached, run_not_funded, account_setup_required; each error carries `fix`: what to do next.

NameTypeReqDescription
acceptAllApplicantsboolean–Accept every eligible applicant automatically instead of vetting creative fit. Eligibility checks and your country/gender restrictions still apply.
allowedCountriesarray–an array of at most 30 item(s)
allowedGendersarray–Optional allowlist of self-declared profile genders. Omit for no gender restriction.
autoModeboolean–Defaults to true. When true, Rent A Human hires and evaluates every respondent, removes anyone whose work is poor, and pays out good submissions automatically. Set false to have every hiring and payo…
idempotencyKeystring–A unique string for this call. Repeat it when you retry, so the call runs once.
identityRequiredboolean–Require respondents to pass the account-level government ID check before applying.
optionsarrayyesThe 2 to 6 things people choose between, each a label and an http(s) link. Labels must differ.
payPerRespondentobjectyesPay per person: at least 50 cents, or 10 for quickVote. Total worker pay must reach 100 cents; including our 18% fee, the run cannot exceed $2,000.
questionstringyesa string matching the pattern \S
quickVoteboolean–true: people tap one of exactly two image options and give no reason. Needs voteMode instant and no video or portfolio. Default false.
requirePortfolioUploadboolean–Require each respondent to upload a portfolio sample with their application.
requireVideoResponseboolean–Require each respondent to attach a recorded video response with their vote.
respondentCountintegeryesa number less than or equal to 100
targetCategoriesarray–Creative categories of respondents to target; omit for no category restriction.
titlestring–a string matching the pattern \S
voteModestring–Defaults to instant. Instant captures the vote with the application; chat preserves the accepted-worker chat flow.
NameTypeReqDescription
costobjectyesWhat the run took from your wallet, on the first call when isRepeat is true. 0 while it waits for money.
isRepeatboolean–true when this answers for a run an earlier call already started; nothing was taken now. Left out otherwise.
runobjectyes–
waitingForobject–Set while the run waits for money. It starts by itself once the wallet covers `needed`; nothing more to call. It ends unstarted after 7 days. Left out otherwise.

No examples provided.

create_webhook_endpoint ~258

Register an https URL that is told about the events you choose, signed with a secret shown only in this reply. At most 5 endpoints. Use for: being told when something happens, for example a new application or message, instead of asking again and again. Not for: seeing the endpoints you have (list_webhook_endpoints); checking an endpoint receives events (test_webhook_endpoint). Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: create_webhook_endpoint({"url":"https://hooks.example.com/rentahuman","events":["application.submitted","message.received"],"description":"Main agent"}) Repeat: A repeat may act again on the thing as it stands then. Read it first. Errors: endpoint_limit_reached; each error carries `fix`: what to do next.

NameTypeReqDescription
descriptionstring–Your own note to tell endpoints apart.
eventsarrayyesEvent names from the allowed list. Send ["*"] for every event.
urlstringyesYour https URL that receives the events. It must reach a public address.
NameTypeReqDescription
endpointobjectyes–
secretstringyesThe signing secret of this endpoint. It is shown only in this reply: store it now to check the signature of each event.

No examples provided.

delete_webhook_endpoint ~214

Remove one of your webhook endpoints. No more events go to it, and its signing secret is gone: an endpoint created again gets a new secret. Use for: stopping events to a URL, making room for another endpoint, or restarting one that was switched off after failures. Not for: seeing your endpoints and their ids (list_webhook_endpoints). Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: delete_webhook_endpoint({"endpointId":"Wk3pQ8vN1sXr6TzL0aBc"}) Repeat: Once this has worked, a repeat does not do it again while that result still stands: it is refused, or answers with what is already there. Errors: only the common ones; each error carries `fix`: what to do next.

NameTypeReqDescription
endpointIdstringyesThe webhook endpoint id.
NameTypeReqDescription
endpointIdstringyes–

No examples provided.

download_picked_photos ~185

Get download links for the picked photos of one hire, as zip files of up to 100 photos each. Links last 15 minutes. Use for: downloading the photos you kept with pick_photos. Not for: looking through the photos one by one (list_hire_photos). Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: download_picked_photos({"hireId":"a3Fh8kQz1WpL0mNc"}) Errors: no_picked_photos, no_submission, no_photo_submission, photos_not_available; each error carries `fix`: what to do next.

NameTypeReqDescription
hireIdstringyesThe hire id, from list_hires or accept_application.
NameTypeReqDescription
expiresAtstringyesWhen the links stop working. Ask again for new ones.
partsarray–One zip file per 100 photos, in order.
pickedCountintegeryesPicked photos in the zip files.

No examples provided.

fund_bounty ~287

Pay for a draft bounty's spots from your wallet. Call with dryRun first to see the total and your balance. Use for: paying for a draft bounty. Not for: paying a worker (pay_hire); adding a spot to a bounty that is already open (update_bounty); paying for one applicant on a bounty paid per hire (accept_application). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: fund_bounty({"bountyId":"b7Kq2mXw9TzR4nLp","dryRun":true}) Repeat: Once this has worked, a repeat does not do it again while that result still stands: it is refused, or answers with what is already there. Errors: wallet_short, bounty_changed, already_funded, not_fundable, payment_processing, spending_limit_reached, account_setup_required; each error carries `fix`: what to do next.

NameTypeReqDescription
bountyIdstringyesThe bounty id, or a pasted rentahuman.ai/bounties/... link.
dryRunboolean–true returns the quote and takes nothing. A dry run shows the price, not whether the charge will be allowed. Default false.
NameTypeReqDescription
bountyobjectyes–
fundingobjectyes–
quoteobject–Set on a dry run. Left out after a real call: a funded reply states no amount. The amount taken is the bounty's price at the moment of the charge, which can differ from an earlier dry run if the boun…

No examples provided.

fund_wallet ~197

Get a card pay link that adds money to your wallet. Nothing is added until the link is paid. Use for: topping up before you fund a bounty or send money. Hand `payUrl` to the person who pays. Not for: checking your balance (get_wallet); paying for a draft bounty (fund_bounty). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: fund_wallet({"amount":{"amount":5000,"currency":"usd"}}) Repeat: A repeat may act again on the thing as it stands then. Read it first. Errors: account_setup_required, funding_unavailable; each error carries `fix`: what to do next.

NameTypeReqDescription
amountobjectyesHow much to add, from $1 to $10,000.
NameTypeReqDescription
amountobjectyes–
expiresAtstring–When the link stops working. Left out: no set time.
machinePaymentobject–Alternative to payUrl for the same deposit. Send body as JSON to url using method before expiresAt. Choose one payment method. Left out unless enabled for this account.
payUrlstringyesOpen this and pay by card. The money is added to your wallet once the payment goes through.

No examples provided.

get_account ~119

See which account and profile your calls act for, and how this call signed in. Use for: checking who you act for before other calls, and whether the account still needs a profile. Not for: your money (get_wallet). Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: get_account({}) Errors: only the common ones; each error carries `fix`: what to do next.

Input schema present but exposes no named parameters.

NameTypeReqDescription
credentialstringyesHow this call signed in: an API key, or hosted sign-in.
profileobject–The profile this call acts for. Left out when the account has no profile yet; then finish setup at `setupUrl`.
setupUrlstring–Where to make a profile. Left out when the account has one.

No examples provided.

get_agent_identity ~74

Get your cryptographic agent identity. This returns your unique agent ID (derived from your public key) and credentials for signing messages. Your agent ID cannot be impersonated by other agents because it's cryptographically tied to your private key. Also shows which named identity is currently active. Call this first before sending messages in supported conversations.

Input schema present but exposes no named parameters.

No output schema declared.

No examples provided.

get_application ~163

Get everything needed to decide on one application: the person, their answers and the bounty's terms. Use for: deciding whether to accept or reject one applicant. Not for: reading every applicant of a bounty (list_applications); reading a hire after the accept (get_hire). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: get_application({"applicationId":"a3Vn8cQe1RuY6wHs"}) Errors: only the common ones; each error carries `fix`: what to do next.

NameTypeReqDescription
applicationIdstringyesThe application id.
NameTypeReqDescription
actionsarray–Tools you may call on this application now. Each tool still checks.
answersarray––
applicantobjectyes–
appliedAtstringyesISO 8601 time with an offset, for example 2026-11-20T09:00:00-08:00.
availabilitystring––
bountyobjectyes–
bountyIdstringyes–
checksobject––
filesarray––
hireIdstring–Set once accepted. Use it with the hire tools.
idstringyes–
linksarray––
messagestring–The cover letter.
replystring–The message sent with the accept or reject.
statusstringyes–

No examples provided.

get_bounty ~176

Get one bounty you posted, with how many people were told about it and applied. Use for: checking the status, spots, settings and outreach of one bounty. Not for: reading who applied (list_applications); reading who you hired (list_hires). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: get_bounty({"bountyId":"b7Kq2mXw9TzR4nLp"}) Errors: only the common ones; each error carries `fix`: what to do next.

NameTypeReqDescription
bountyIdstringyesThe bounty id, or a pasted rentahuman.ai/bounties/... link.
NameTypeReqDescription
actionsarray–Tools you may call on this bounty now. Each tool still checks.
aiobject––
applicationsobjectyes–
bonusesarray––
categorystring––
checksobject––
countriesobject––
createdAtstringyesISO 8601 time with an offset, for example 2026-11-20T09:00:00-08:00.
descriptionstringyes–
fundingobjectyes–
idstringyes–
imageUrlsarray––
locationobject––
messagesobject––
outreachobject–How many workers were told about the bounty and applied. Left out only when you are a teammate of the organization that posted it: the account that posted the bounty always sees these numbers.
payobjectyes–
paymentobjectyes–
proofobjectyes–
requirementsarray––
scheduleobject––
skillsarray––
spotsobjectyes–
startsAtstring–ISO 8601 time with an offset, for example 2026-11-20T09:00:00-08:00.
statusstringyes–
timeZonestring–IANA time zone for every time on this bounty.
timersobject––
titlestringyes–
updatedAtstringyesISO 8601 time with an offset, for example 2026-11-20T09:00:00-08:00.
urlstringyes–

No examples provided.

get_cake_order ~224

Get one cake order you sent: its status, who it goes to, the delivery photos and what it cost. Use for: checking where a cake order stands after send_cake. It only reads the stored order: it never asks the vendor, sends the order again or moves money. Not for: ordering a cake (send_cake). Needs: an API key. Hosted sign-in cannot call this yet. Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: get_cake_order({"orderId":"cake_3f9a0c1e2b4d5f6a7b8c9d0e1f2a3b4c5d6e7f80"}) Errors: only the common ones; each error carries `fix`: what to do next.

NameTypeReqDescription
orderIdstringyesThe order id send_cake returned.
NameTypeReqDescription
chargeobjectyesWhat the order took from your wallet: price + fee.
costobjectyes–
orderobjectyes–

No examples provided.

get_cake_quote ~279

Ask whether a cake can be delivered to an address, and what each size costs, per cake. Use for: checking delivery and prices before sending a cake. Nothing is ordered or paid. Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: get_cake_quote({"address":"1 Market St","city":"San Francisco","state":"CA","zip":"94105"}) Errors: quote_unavailable; each error carries `fix`: what to do next.

NameTypeReqDescription
addressstring–Street and number. Send it with city, state and zip.
address2string–Suite, floor or unit.
citystring–a string at most 200 character(s) long
quantityinteger–How many cakes, 1 to 100. Default 1. It can change what sizes are offered; prices stay per cake.
statestring–a string at most 100 character(s) long
websiteUrlstring–A company website. The bakery finds the address. Send this, or a full address.
zipstring–a string at most 20 character(s) long
NameTypeReqDescription
isDeliverableboolean–Left out when the address cannot get a cake.
notesstring–What the bakery says about larger orders, when it says it.
sizesarray––

No examples provided.

get_cake_status ~71

Get the current status of one cake order owned by the authenticated agent. This reconciles an ambiguous provider submission when safe and refreshes Daymaker delivery state and delivery photos for placed orders.

NameTypeReqDescription
order_idstringyesa string matching the pattern ^cake_[a-f0-9]{40}$

No output schema declared.

No examples provided.

get_conversation ~197

Get one of your conversations with a worker and its first 100 messages, oldest first. When `messages.nextCursor` is set, pass it to list_messages for the rest. Use for: opening a conversation: who it is with, whether you can send now, and what was said. Not for: finding a conversation (list_conversations); reading past the first 100 messages (list_messages). Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: get_conversation({"conversationId":"c7Rk2mQw9TzL4nXp"}) Errors: only the common ones; each error carries `fix`: what to do next.

NameTypeReqDescription
conversationIdstringyesThe conversation id, from list_conversations or the conversationId of a hire.
NameTypeReqDescription
conversationobjectyes–
messagesobject–The first 100 messages, oldest first. With `nextCursor`, read the rest with list_messages.

No examples provided.

get_hire ~196

Get one hire: the worker, the money, what they submitted and what you can do next. Use for: reviewing the work of one worker, how many photos wait for you, the dispute and the latest message. Not for: listing every worker on a bounty (list_hires); looking through the photos (list_hire_photos); finding work that waits for review (list_submissions). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: get_hire({"hireId":"a3Fh8kQz1WpL0mNc"}) Errors: only the common ones; each error carries `fix`: what to do next.

NameTypeReqDescription
hireIdstringyesThe hire id, from list_hires or accept_application.
NameTypeReqDescription
acceptedAtstring–When you accepted the worker. Left out only on an older hire that did not store it.
actionsarray–Tools you may call on this hire now. Each tool still checks.
attendanceobject–Where the worker is for this round of work, from their phone. Left out on a bounty that is not in person, or before they start sharing. Never coordinates.
bountyIdstringyes–
conversationIdstring––
costobject–What this hire costs you as its held money stands now: pay plus service fee. Left out when nothing is held: the fee of a pay-later hire is set when you pay.
disputeobject––
dueobject––
endedAtstring–ISO 8601 time with an offset, for example 2026-11-20T09:00:00-08:00.
endedReasonstring––
idstringyesThe hireId.
isLatestMessageHiddenboolean–true when the chat could not be shown because the money record this hire points at could not be proven: a message may exist. Left out otherwise.
latestMessageobject–The newest chat message you can see, once the worker has submitted work. Left out before that, when there is none, and when isLatestMessageHidden is true.
paidAtstring–ISO 8601 time with an offset, for example 2026-11-20T09:00:00-08:00.
payobject–The worker's pay for this hire, bonuses included: base pay plus the most the bonuses can pay until they are decided, then base pay plus the bonuses earned. Left out only when it cannot be read: every…
payStatusstringyesWhere the money stands. paid means it was released to the worker and left your side for good; it does not say the worker's balance was credited yet.
statusstringyesoffered: waiting for the worker to confirm. active: working, or fixing a redo. submitted: work waits for your review, or the work is done and its pay waits for you to release or pay it. paid: the pay…
submissionobject–The work the worker sent for this hire. Left out before they submit; a draft is never shown.
submittedAtstring–ISO 8601 time with an offset, for example 2026-11-20T09:00:00-08:00.
workerobject–The hired worker. id or name is left out only on an older hire that did not store it.

No examples provided.

get_hire_location_track ~187

Get where one hired worker was on an in-person bounty that requires precise location: the site, the latest position and up to 1,000 recorded positions. Use for: checking where a worker was during this round of work, when they share a precise location. Not for: seeing who is on site across a bounty (list_hires). Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: get_hire_location_track({"hireId":"a3Fh8kQz1WpL0mNc"}) Errors: location_not_shared, no_location_track; each error carries `fix`: what to do next.

NameTypeReqDescription
hireIdstringyesThe hire id, from list_hires or accept_application.
NameTypeReqDescription
deleteAfterDaysintegeryesPositions are deleted this many days after they are recorded.
gapsarray–Silences between recorded positions.
hireobjectyes–
isTruncatedboolean–More positions were recorded than points holds.
latestobject––
pointsarray–Recorded positions only, newest first, at most 1,000. Nothing is filled in.
sharingEndedAtstring–When the worker stopped sharing a precise location. Left out while sharing is on.
siteobjectyes–

No examples provided.

get_human ~172

Get one person’s full public profile: bio, languages, hours, rate, rating and paid track record. Use for: deciding whether to hire someone you found. Not for: finding people (list_humans); reading what others wrote about them (list_reviews). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: get_human({"humanId":"h2Lw9QxT4mZp7Rk"}) Errors: only the common ones; each error carries `fix`: what to do next.

NameTypeReqDescription
humanIdstringyesThe id from list_humans. Their username also works.
NameTypeReqDescription
accountTypestringyes–
availabilityobject–Hours they say they can work each day, HH:mm in `timezone`.
avatarUrlstring–May be left out while their profile is incomplete.
biostring––
certificationsarray––
createdAtstringyesISO 8601 time with an offset, for example 2026-11-20T09:00:00-08:00.
educationstring––
expertisearray––
headlinestring––
hourlyRateobject–What they ask per hour. Left out when their rate is in another currency.
idstringyes–
isAvailableboolean––
isVerifiedboolean––
languagesarray––
locationobject––
namestring–May be first name and last initial while their profile is incomplete. Left out when they gave none.
paidJobCountinteger–Jobs they were paid for through the platform. Left out when it could not be counted.
photoUrlsarray––
ratingobjectyes–
repeatPosterCountinteger–Posters who paid them for two or more jobs. Left out when it could not be counted.
skillsarray––
timezonestring––
urlstringyes–
usernamestring––
verifiedLanguagesarray–Languages proven by a speaking check.
withdrawalCountinteger–Jobs they left after starting. Left out when it could not be counted.
workHistorystring––

No examples provided.

get_humanization ~173

Get one order for a human rewrite of a text: its status, writer and the rewritten text once it is sent. Use for: reading the rewritten text and where a humanization order stands. Not for: its spots, outreach or settings as a bounty (get_bounty). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: get_humanization({"humanizationId":"h7Kq2mXw9TzR4nLp"}) Errors: only the common ones; each error carries `fix`: what to do next.

NameTypeReqDescription
humanizationIdstringyesThe id create_humanization returned.
NameTypeReqDescription
humanizationobjectyes–

No examples provided.

get_qa_run ~194

Get one QA run you started: where the testing stands, what it cost, the report and its findings, and what each tester sent. Use for: checking a QA run’s progress and reading its findings and each tester’s written account and evidence. Not for: finding a run id (list_qa_runs); reading a taste run (get_taste_run); checking your wallet balance (get_wallet). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: get_qa_run({"runId":"qa_run_7Np3sD8"}) Errors: only the common ones; each error carries `fix`: what to do next.

NameTypeReqDescription
runIdstringyesThe run id the create call or the list answered with.
NameTypeReqDescription
isTestersTruncatedboolean–True when the run has more testers than shown.
runobjectyes–
testersarray–Newest first, at most 200.

No examples provided.

get_taste_run ~201

Get one taste run you started: its status, the live vote count per option, what it cost, and the report once it is closed. Use for: checking where a taste run stands and reading its result. While it is collecting, read it again in 20 to 30 minutes. Not for: reading a QA run (get_qa_run); checking your wallet balance (get_wallet). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: get_taste_run({"runId":"taste_8Hc2pQ1vLm"}) Errors: only the common ones; each error carries `fix`: what to do next.

NameTypeReqDescription
runIdstringyesThe run id create_taste_run answered with, or one from list_taste_runs.
NameTypeReqDescription
runobjectyes–

No examples provided.

get_wallet ~132

Get your wallet balance, your spending limits, and whether you can add money by card now. Use for: checking what you can spend before you fund a bounty or send money, and reading your limits. Not for: money that came in or went out (list_wallet_transactions). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: get_wallet({}) Errors: account_setup_required; each error carries `fix`: what to do next.

Input schema present but exposes no named parameters.

NameTypeReqDescription
balanceobjectyesYour stored wallet balance, earnings kept in the wallet included. Sends and bounty funding are checked against it.
controlsobject––
fundingobject––

No examples provided.

get_webhook_delivery ~224

Get one delivery to your endpoints: where it stands, the exact body that was signed and sent, and each try with the status your URL gave. Use for: seeing whether a test or real event reached your URL, and why it failed if it did not. Not for: finding deliveries and their ids (list_webhook_deliveries). Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: get_webhook_delivery({"deliveryId":"3f9c2a7b1e5d8c4f0a6b2e9d7c3f1a5b8e4d0c6a2f9b7e3d1c5a8f4b0e6d2c9a"}) Errors: only the common ones; each error carries `fix`: what to do next.

NameTypeReqDescription
deliveryIdstringyesThe delivery id.
NameTypeReqDescription
attemptCountintegeryesTries made so far.
attemptsarray–The last 10 tries at most, oldest first.
createdAtstringyesISO 8601 time with an offset, for example 2026-11-20T09:00:00-08:00.
deliveredAtstring–When your URL accepted it. Left out until then.
endpointIdstringyes–
eventIdstringyesThe id of the event. A resent event keeps its id.
eventTypestringyes–
failedAtstring–When the last try failed for good. Left out until then.
idstringyesThe delivery id to call get_webhook_delivery with.
lastErrorstring–Why the last try failed. Left out when none failed.
maxAttemptsintegeryes–
payloadstringyesThe exact JSON text that was signed and sent to your URL, unchanged, so its signature checks against it. Parse it to read the event.
responseStatusinteger–HTTP status your URL gave on the last try. Left out before the first answer.
statusstringyespending: waiting to be sent. claimed: being sent now. delivered: your URL answered 2xx. retry: it failed and will be sent again. failed: no more tries.

No examples provided.

invite_human ~282

Ask one person, by id, to apply to your open bounty. Use for: a poster who already knows which person they want. Inviting the same person to the same bounty again sends nothing new. Not for: telling many matching workers at once (invite_more_applicants); finding the person’s id (list_humans); seeing whether they applied (list_applications). Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: invite_human({"bountyId":"b7Kq2mXw9TzR4nLp","humanId":"h2Lw9QxT4mZp7Rk"}) Repeat: A repeat with the same input, and the same key if you sent one, does not act twice: it returns the same result, finishes the first call, or is refused. Errors: not_found, not_open, worker_unavailable, already_applied; each error carries `fix`: what to do next.

NameTypeReqDescription
bountyIdstringyesThe bounty id, or a pasted rentahuman.ai/bounties/... link.
humanIdstringyesThe human id, as `id` in list_humans.
NameTypeReqDescription
invitationobjectyes–
isEmailSentboolean–True only when the email about this invitation is confirmed sent. Left out when it is not confirmed: they turned off bounty email, the email could not be sent, or its delivery was not recorded. The i…
isRepeatboolean–True when they were already invited to this bounty. Nothing new was stored or sent.

No examples provided.

invite_more_applicants ~207

Tell another wave of matching workers about an open bounty. Once per bounty per day. Use for: a bounty that needs more applicants. The reply says how many workers were put in line in this call, not how many messages arrived. Not for: changing the bounty so more people fit (update_bounty). Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: invite_more_applicants({"bountyId":"b7Kq2mXw9TzR4nLp"}) Repeat: A repeat may act again on the thing as it stands then. Read it first. Errors: forbidden, not_open, invited_recently; each error carries `fix`: what to do next.

NameTypeReqDescription
bountyIdstringyesThe bounty id, or a pasted rentahuman.ai/bounties/... link.
NameTypeReqDescription
bountyobjectyes–
outreachobjectyes–

No examples provided.

list_api_keys ~156

List the API keys of your account, newest first. The secret of a key is never shown again. Use for: finding the id of a key to revoke, and seeing which keys are active and when they were last used. Not for: making a new key (create_api_key); turning a key off (revoke_api_key). Needs: an API key. Hosted sign-in cannot call this yet. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: list_api_keys({}) Errors: only the common ones; each error carries `fix`: what to do next.

Input schema present but exposes no named parameters.

NameTypeReqDescription
dataarray–Every key of this account, revoked ones too, newest first.

No examples provided.

list_applications ~309

List who applied to one of your bounties, best matches first. Use for: reviewing applicants, or finding every hire including ended ones (status: accepted). Not for: reading one application with the profile and the bounty terms (get_application); reading what hired workers did (list_hires). Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: list_applications({"bountyId":"b7Kq2mXw9TzR4nLp","status":"pending","limit":20}) Errors: only the common ones; each error carries `fix`: what to do next.

NameTypeReqDescription
bountyIdstringyesThe bounty id, or a pasted rentahuman.ai/bounties/... link.
cursorstring–The nextCursor of the previous page. Leave it out for the first page.
limitinteger–Rows wanted. Each call reads one stretch of the applications, for every sort and status, so a page can hold fewer rows than `limit`, or none, while `nextCursor` says more remain; only a page without…
sortstring–best: matches ranked again each call; pages may repeat or miss rows. newest: stable, newest first. Unreadable records are omitted.
statusstring––
NameTypeReqDescription
dataarray––
nextCursorstring––

No examples provided.

list_blocked_humans ~153

Everyone you blocked from applying to your bounties, newest first, with your reason. Use for: checking whether you blocked someone, or finding whom to unblock. Not for: letting someone apply again (unblock_human); people you like to work with (list_preferred_humans). Needs: an API key. Hosted sign-in cannot call this yet. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: list_blocked_humans({}) Errors: account_setup_required; each error carries `fix`: what to do next.

Input schema present but exposes no named parameters.

NameTypeReqDescription
dataarray–Everyone you blocked, newest first, up to 500.

No examples provided.

list_bounties ~282

List your own bounties, newest first: the ones you or your agent posted, and your organization's; search_bounties finds open bounties to apply to. A page can hold fewer rows than `limit`, or none; keep going until a page has no `nextCursor`. Not listed: a bounty priced in another currency, or one the website's list also hides from you. A bounty you post while paging is not in that walk; start again to see it. Use for: finding a bounty you can manage, or every bounty in one status. Not for: reading one bounty in full (get_bounty); reading who applied (list_applications). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: list_bounties({"status":"open","limit":20}) Errors: only the common ones; each error carries `fix`: what to do next.

NameTypeReqDescription
cursorstring–The nextCursor of the previous page. Leave it out for the first page.
limitinteger–Rows wanted. A page can hold fewer, or none; only a page without nextCursor is the end.
statusstring––
NameTypeReqDescription
dataarray––
nextCursorstring––

No examples provided.

list_conversations ~312

List your conversations with workers, newest activity first: who, the subject, unread messages, and whether you can send now. Keep going until a page has no `nextCursor`. Use for: finding the conversation with a worker, or the ones with messages you have not read. Not for: reading the messages of one conversation (get_conversation); reading what a hired worker did (get_hire). Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: list_conversations({"isUnread":true,"limit":20}) Errors: only the common ones; each error carries `fix`: what to do next.

NameTypeReqDescription
bountyIdstring–The bounty id, or a pasted rentahuman.ai/bounties/... link.
cursorstring–The nextCursor of the previous page. Leave it out for the first page.
hasRepliesboolean–true: only conversations with more than one message. Left out or false: every conversation.
isUnreadboolean–true: only conversations with messages from the worker you have not read. Left out or false: every conversation.
limitinteger–Rows wanted. A page can hold fewer, or none; only a page without nextCursor is the end.
statusstring––
subjectstring–Only conversations with exactly this subject.
NameTypeReqDescription
dataarray––
nextCursorstring––

No examples provided.

list_hire_photos ~250

List a hired worker’s bounty photos in the order sent. Continue until there is no nextCursor. Use for: looking through the photos before you pick them with pick_photos. Not for: how many photos are in each state (get_hire); downloading the picked photos (download_picked_photos). Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: list_hire_photos({"hireId":"a3Fh8kQz1WpL0mNc","filter":"to_review"}) Errors: no_submission, no_photo_submission; each error carries `fix`: what to do next.

NameTypeReqDescription
cursorstring–The nextCursor of the previous page. Leave it out for the first page.
filterstring–Only photos in this state. Default: all.
hireIdstringyesThe hire id, from list_hires or accept_application.
limitinteger–Rows wanted. A page can hold fewer, or none; only a page without nextCursor is the end.
NameTypeReqDescription
dataarray––
nextCursorstring––

No examples provided.

list_hires ~406

List the workers you hired on one bounty, by when they applied, newest first, with where each one stands and who is on site. Each call reads one stretch of the applications, so a page can hold fewer rows than `limit`, or none, while `nextCursor` says more remain; keep going until a page has no `nextCursor`. Each hire is read when the list reaches it, so one that changes while you page is still listed once. An applicant you accept after the list has passed them is not in that walk; start again to see them. Use for: seeing every worker on a bounty, their money and what waits for you. Not for: reading one hire in full, with the work (get_hire); finding work that waits for review (list_submissions); reading who applied and is not hired (list_applications). Money: US cents. Replies omit a field's sole empty value. If null is allowed, only null is omitted; false, [] and a present group stay. A present empty group is {}. Field descriptions explain unavailable values. Example: list_hires({"bountyId":"b7Kq2mXw9TzR4nLp","status":"active","limit":20}) Errors: only the common ones; each error carries `fix`: what to do next.

NameTypeReqDescription
bountyIdstringyesThe bounty id, or a pasted rentahuman.ai/bounties/... link.
cursorstring–The nextCursor of the previous page. Leave it out for the first page.
limitinteger–Rows wanted. Each call reads one stretch of the applications and builds at most `limit` hires from it, so a page can hold fewer rows than `limit`, or none, while `nextCursor` says more remain; only a…
statusstring–Only hires in this status. Default: every hire.
NameTypeReqDescription
dataarray––
nextCursorstring––

No examples provided.

Install

How do I install the RentAHuman MCP server?

RentAHuman runs locally as an npm package, launched with npx -y rentahuman-mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

npm · rentahuman-mcp

# add to Claude Code
claude mcp add rentahuman-ai-rentahuman -- npx -y rentahuman-mcp
Common questions

What is the RentAHuman MCP server?

RentAHuman is an MCP server listed in the public MCP registry as io.github.rentahuman-ai/rentahuman. Hire humans for digital microtasks, research, QA, data evaluation, and physical-world tasks. This page covers its npm package (rentahuman-mcp).

Is the RentAHuman MCP server safe to use?

RentAHuman scores 67 out of 100 on VerifyMCP. We recorded 4 known advisories against it as of 21 November 2026. It declares no install or post-install scripts. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

What tools does the RentAHuman MCP server expose?

RentAHuman exposes 69 tools: accept_application, block_human, cancel_bounty, cancel_hire, create_api_key, and 64 more. Their descriptions and schemas cost roughly 20,252 tokens of context every time the server is loaded.

Is the RentAHuman MCP server still maintained?

RentAHuman is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

What licence is the RentAHuman MCP server under?

RentAHuman declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

What did VerifyMCP test on the RentAHuman MCP server?

VerifyMCP ran 15 checks on RentAHuman, covering Malware scan, Known CVEs, Install scripts, Dependency health and 11 more. One item could not be completed, a gap on our side rather than a finding about RentAHuman: Security disclosure. Endpoint Security and Transport & Reachability do not apply to an npm package, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.