SameDayDesk agent tools

com.samedaydesk由 samedaydesk.com 發布官方發布者v1.2.0更新於 Oct 31, 2026

Readiness checks, a paid Fix Pack, TaskMarket reads, and caller-declared funnel evidence.

已驗證Streamable HTTP可網頁執行網頁搜尋與抓取開發者工具商業與電商

概覽

AI 產生的概覽

檢查網站能否被 AI 搜尋引擎與代理程式發現,提供付費修復包,並讀取公開 TaskMarket 任務資料。

可以做什麼
  • 為網站的 AI 搜尋可見性與代理程式可發現性評分,並逐項給出修正建議
  • 憑授權碼產生付費的 AI 就緒修復包(JSON-LD、robots.txt、網站地圖、meta 修正)
  • 瀏覽並依條件篩選公開 TaskMarket 任務,追蹤單一任務的狀態、提交與獎勵
  • 規劃附支出上限的受限 TaskMarket 委託內容
  • 將呼叫端提供的觀測資料包投影為漏斗決策
適用情境
適合需要快速評估網站在 AI 爬蟲與代理程式眼中的可見性、需要可直接貼上的結構化資料與爬蟲修正方案,或希望以唯讀方式查看公開 TaskMarket 任務清單與任務詳情時使用。
執行需求
遠端 streamable HTTP 端點;未宣告驗證、環境變數或標頭。付費修復包需要從另行結帳流程取得的 license 值。TaskMarket 相關工具會讀取公開鏈上中繼資料與官方唯讀 API,因此需要網路存取。
安裝前請注意
其中一個工具需付費:修復包售價 39 美元,需要外部結帳取得的 license。TaskMarket 工具讀取公開鏈上資料與任務描述,任務描述以不受信任的文字回傳。委託規劃工具只產生附支出上限的內容,不授權付款;追蹤工具無法建立、接受或退款任務。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 SameDayDesk agent tools,將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Web executable,透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式,或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南,先向你展示發布者、端點和登入方式,經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/com.samedaydesk%2Ftask-evidence/install.md,並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前,先告訴我它的發布者、端點和登入方式,等我確認。不要把 API 金鑰寫進命令或我會提交的檔案;問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "samedaydesk-agent-tools": {
      "type": "http",
      "url": "https://samedaydesk.com/mcp"
    }
  }
}

README

SameDayDesk (public entry)

Public helpers for free discovery and offline result reuse. Node 22.x. The commands below use plain ESM under tools/; no npm install is required.

Start with the examples below, then open the helper documentation for options and error handling.

Docs (in-repo)

DocPurpose
tools/presence/FOR-AGENTS-COLD-READ.mdFree discovery when apex TLS/connect fails
tools/presence/REGISTRY-CONSUMER.mdMCP Registry version=latest pitfall
tools/result-reuse/README.mdOffline source → task-memory observation
tools/offer-routing/README.mdTask → existing SameDayDesk/Neomorphic offer
tools/offer-routing/capability-limits-matrix.jsonMachine-readable capability/limits matrix

Live free catalogs (no pay): https://agents.samedaydesk.com/llms.txt and https://agents.samedaydesk.com/.well-known/skills/index.json. Apex https://samedaydesk.com/for-agents may TLS/connect-fail; cold-read falls back to those alternates.

Free / live vs offline fixtures

ModeNetworkPaymentHow
Offline fixture cold-readnonenonepreferFixture: true → outcome: "offline_fixture"
Live cold-readfree HTTP onlynonedefault resolve → often alternate_live when apex fails
Result-reuse preview/exportnone (local JSON)nonefixtures + --opt-in to write

These helpers do not require a purchase. Paid extract/batch routes are separate.

Exact commands (clean checkout)

From the repository root (no install):

1. Offline cold-read (no network)

bash
node --input-type=module -e 'import {resolveForAgentsColdRead} from "./tools/presence/for-agents-cold-read.mjs"; console.log(JSON.stringify(await resolveForAgentsColdRead({preferFixture:true}), null, 2))'

Expected: outcome: "offline_fixture", paid: false, liveObserved: false.

2. Live cold-read (free HTTP only; optional)

bash
node --input-type=module -e 'import {resolveForAgentsColdRead} from "./tools/presence/for-agents-cold-read.mjs"; console.log(JSON.stringify(await resolveForAgentsColdRead(), null, 2))'

Expected when apex transport fails but agents.* respond: outcome: "alternate_live", paid: false.

3. Result-reuse preview and export on in-repo fixtures

bash
PILOT_EXAMPLE_DIR="$(mktemp -d)"NOW="$(date -u +%Y-%m-%dT%H:%M:%SZ)"node tools/result-reuse/cli.mjs preview \  --input tools/result-reuse/fixtures/accepted-page-change.json \  --task-id vendor-watch --subject vendor-page-result --sequence 1 --clock "$NOW"
node tools/result-reuse/cli.mjs export \  --input tools/result-reuse/fixtures/accepted-page-change.json \  --task-id vendor-watch --subject vendor-page-result --sequence 1 --clock "$NOW" \  --opt-in --out "$PILOT_EXAMPLE_DIR/reuse-observation.json"printf 'Saved example in %s\n' "$PILOT_EXAMPLE_DIR"

Record-kind fixture:

bash
PILOT_EXAMPLE_DIR="$(mktemp -d)"NOW="$(date -u +%Y-%m-%dT%H:%M:%SZ)"node tools/result-reuse/cli.mjs export \  --input tools/result-reuse/fixtures/accepted-record-report.json \  --task-id vendor-watch --subject record-export --sequence 1 --clock "$NOW" \  --opt-in --out "$PILOT_EXAMPLE_DIR/reuse-record-observation.json"printf 'Saved example in %s\n' "$PILOT_EXAMPLE_DIR"

Expected written observation: schema: "neomorphic.task-memory.observation.v1"; export requires --opt-in plus caller clock/task/subject/sequence.

Task → existing offer (cold selection)

When the job is known, route with the matrix before paying or treating a sample as live:

bash
node tools/offer-routing/route-job.mjs tools/offer-routing/fixtures/complete-issue-discussion.job.json

Expected: ok: false, selected: null, exit code 2 and complete_issue_acquisition_unavailable. Neither paid /extract nor task-kit packaging acquires a complete discussion. This matrix does not advertise the pending issue-evidence acquisition pack. For historical fixture composition use fixtures/supplied-issue-brief.job.json under tools/offer-routing/. The MoltJobs SDK pack runs a disposable local HTTP rehearsal; its shipped runner refuses --live, even with credentials. Hosted archives are downloads, not hosted job execution. Routing does not evaluate acceptance criteria or authorize payment.

Verify the cited examples

bash
npm run test:public-entry

That script runs the offline cold-read and fixture export commands cited above (no network, no payment). Existing focused suites: npm run test:presence, npm run test:result-reuse, npm run test:offer-routing.

來源:README.md,提交 e27a318

工具

7

check_ai_readiness

Check whether a website is visible to AI search engines (ChatGPT, Perplexity, Claude, Google AI Overviews). Scores AI-crawler access, JSON-LD structured data, title/meta, Open Graph, sitemap, and llms.txt, and returns a 0-100 score, a letter grade, and a specific fix for each gap. Free.

輸入參數url必填

check_agent_readiness

Free. Check whether agents can discover and call a public website. Fetches discovery files, follows MCP links named by those files, and runs only initialize, tools/list, and one unknown-tool call. Never executes a real tool. Returns a 0-100 score, every check with an evidence URL, and the top fixes. No license.

輸入參數host必填

generate_complete_fix_pack

PAID. Returns the complete, ready-to-paste AI-readiness Fix Pack for a site: tailored Organization + FAQPage JSON-LD, an AI-crawler robots.txt, a sitemap, and title/meta/Open Graph fixes. Requires a `license` — the checkout-session id you receive after buying the $39 Fix Pack at https://buy.stripe.com/8x24gA0xA9DF9dd13YeZ20h (after paying you're shown your license code). Without a valid license it returns purchase instructions plus a free starter pack.

輸入參數url必填license

plan_taskmarket_delegation

Prepare a bounded TaskMarket delegation for research, coding, data collection, benchmarking, or verification. Requires chain_id, token_address, deliverable, reward, deadline, and an explicit maximum-spend ceiling. Reads public chain metadata. Returns the canonical TaskMarket API payload and approval summary without creating a task, holding a wallet, or spending funds.

輸入參數modetagsrequest必填chain_id必填deliverable必填reward_usdctoken_address必填deadline_hours必填max_spend_usdcreward_base_unitsacceptance_criteriamax_spend_base_units

browse_taskmarket_tasks

Browse current public TaskMarket inventory through the official read API. Filter by lifecycle status, mode, tag, text, and minimum reward. Amounts carry their actual chain/token; unknown decimals stay raw. Task descriptions are returned as untrusted text and are never executed.

輸入參數tagmodelimitsearchstatuschain_idmin_reward_usdc

track_taskmarket_task

Track one public TaskMarket task through the official read API. Returns status, deadline, submissions, artifact hashes, canonical awards, pending actions, and the next authorization boundary. It cannot create, accept, reject, rate, or refund work.

輸入參數task_id必填

project_funnel_evidence

Free. Project one declared source-separated observation packet into the accepted funnel decision. Caller observations stay declared. Does not fetch URLs, query customers, store the packet, or spend.

輸入參數packet必填

版本歷史

1
  1. v1.2.0最新Oct 10, 2026

評分

已滿足適用檢查項的 90%

該評分衡量此收錄條目的完整程度,依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價,也不表示伺服器安全或目前可用:可連通性是帶日期的檢查結果,評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分

    請在儲存庫或註冊中心條目中宣告授權條款。

其他 Samedaydesk 整合