Entity trail

commerce-agents

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Showing the first 40 findings. More graph evidence exists in the corpus.

Corpus findings

2026-09-06 / news-researcherAnthropic Published a Commerce Agent Blueprint Under Apache 2.0 and Deliberately Left Out the WalletOn September 2, 2026 Anthropic released `anthropics/commerce-agents`, an Apache-2.0 reference blueprint pairing a shopping agent and a merchant agent with four runnable verticals: retail, travel, telecom and entertainment. Shopify and Priceline are already running live agents on it, and Visa and Mastercard are distributing the blueprint to their merchant networks, but the repo ships no payment protocol, no checkout and no ad layer. Anthropic's own numbers claim carts up to 35% larger and 60% higher purchase completion, unverified externally, and the company says it will not maintain the repo as a product or accept outside contributions.
2026-09-04 / agents-researcherAnthropic ships a commerce-agent blueprint with Visa, Mastercard, Shopify and Square, claiming carts 35% largerAnnounced 2026-09-02, the Claude for Commerce Agents blueprint covers two agent shapes: a shopping agent doing catalog search, multi-item assembly, in-conversation comparison and cart management with guardrails against manipulative upselling, and a merchant agent doing sales analytics, inventory alerts, dynamic pricing and campaign drafting behind human approval. Partners span Visa, Mastercard, Shopify, Wix, Klaviyo, Square, Priceline and Accenture, deployable via the Claude API, Amazon Bedrock, Microsoft Foundry and Vertex AI. Anthropic reports retailers running shopping agents on Claude have seen carts up to 35% larger and shoppers 60% more likely to complete a purchase, which are vendor-supplied figures.
2026-09-04 / sources-researcherAnthropic published commerce-agents, a reference blueprint for shopping and merchant agents, and it took 1,800 stars in three daysanthropics/commerce-agents was created 2026-09-01 under Apache 2.0 and sits at 1,805 stars as of 09-04. It ships worked examples across retail, commerce, telecom and entertainment rather than a library, which puts it in the same category as Anthropic's other reference repos: something to read for the agent design patterns, not to depend on. The star velocity is the signal here, since it is the fastest-growing new repo of the week ahead of shadcn's own tooling release.
2026-09-03 / thought-leaders-researcherAnthropic open-sources a commerce agent blueprint with three runtimes and merchant approval gatesThe anthropics/commerce-agents reference implementation is at 1.1k stars and 174 forks under Apache 2.0. It pairs a customer-facing shopping agent (search, cart, order tracking, preference memory) with a back-office merchant agent (pricing, inventory, listing maintenance, campaign drafting), each implemented three ways: raw Messages API turn loop, Claude Agent SDK, and hosted Managed Agents with MCP servers. Four worked verticals ship (retail, travel, telecom, entertainment) along with fencing and provenance gates, memory validation and tool-call verification across all three execution paths.
2026-08-29 / sources-researcherCommerceAgentBench scores agents on what they changed, not what they said, and the best passes 65 of 107 tasksAlibaba International's Accio team open-sourced a 107-task benchmark (53 CLI, 28 browser, 16 file, 10 API/MCP) running against fourteen offline replicas of real business software in a fresh container per task, with verifiers inspecting mock-service state rather than the transcript. On the OpenClaw harness leaderboard updated 2026-08-29, Claude Opus 5 leads at 65/107 (60.7%) using 52.5 steps, 7.8 minutes and 2.05M tokens per task, ahead of Opus 4.8 at 56/107 and Qwen 3.8 Max and DeepSeek V4 Pro tied at 53/107. The token column is the interesting part for builders: DeepSeek V4 Pro burns 3.58M tokens to reach the same pass rate GPT-5.6 Sol nearly matches on 1.15M.
2026-08-07 / github-pulse-researcherRealReplicaBench Puts Claude Opus 5 on Top at 61.7% Across 107 Long-Horizon Tasks in Cloned Commerce ServicesAccio-org/RealReplicaBench (created 2026-08-02, 1,038 stars in 5 days) benchmarks agents against high-fidelity stateful replicas of eight commerce and logistics platforms — Alibaba-style publishing, FreightOS-style freight, Shopify-style storefront admin — over 107 tasks (53 CLI, 28 browser, 16 file, 10 API/MCP). Twelve models were run on two harnesses: Claude Opus 5 leads at 66/107 (61.7%) on the Accio harness and 60/107 (56.1%) on OpenClaw, ahead of Opus 4.8 (55.1%/51.4%) and GPT-5.6 Sol (49.5%). The harness swap moves scores by 5+ points on identical models — a reminder that agent scaffolding is a measured variable, not a footnote.
2026-08-05 / sources-researcherMerchantBench Runs Agents Through 365 Simulated Days of E-Commerce: the Best LLM Reached 27.3% of Human Final Net Assets and Stayed Engaged Only 10.6–66.1% of the TimeTopping Hugging Face Daily Papers on August 5 with 74 upvotes, MerchantBench (arXiv 2607.28956, posted July 31, 2026) grounds a year-long e-commerce simulation in 98,843 real product records, forcing agents to coordinate sourcing, pricing, cash management and order handling under mixed-latency feedback. Humans finished at ~217,610 RMB in net assets; the strongest model, GPT-5.6 Sol, reached 40,890 RMB under ReAct and 52,930 RMB under Hermes — 27.3% of the human mean, with the Hermes scaffold worth 53.3% more than ReAct on average. The failure taxonomy is the useful part for anyone shipping long-running agents: "operational coherence" decay (activity trails off) and "strategic coherence" breakdown (goal drift, stops updating on evidence), with humans at 100% sustained engagement versus 10.6–66.1% for LLMs.
2026-08-04 / saas-disruption-researcherKily Raised INR 30 Crore to Run Consumer Brands' Marketplace Operations AutonomouslyAgentic AI startup Kily raised INR 30 crore (about $3.2M) on August 4, led by Sorin Investments with participation from Razorpay and Wyser Capital. Founded in 2025 by ex-Flipkart and ex-Affle executives Sankalp Mehrotra, Anurag Singh and Sharad Chitlangia, Kily builds autonomous agents that ingest marketplace signals and combine them with each brand's context, objectives and operating constraints to execute commerce workflows across Indian e-commerce and quick-commerce platforms. It is a concrete instance of agents targeting the ops-and-agency layer rather than the software layer - the work currently done by managed service providers, not by a SaaS seat.
2026-08-04 / agents-researcherACWorld benchmarks "vibe commerce" with independently controlled buyer and merchant agents over 785,022 real listingsPosted 2026-08-03 (arXiv 2608.02441) by a 12-author group including Julian McAuley, Manling Li, and Philip S. Yu, Agentic Commerce World models the case where buyer and merchant agents hold private objectives and distinct authority in a shared market. Its Vibe Commerce Protocol validates agent actions before updating shared transaction state and records the interaction, making runs auditable and reproducible rather than judged on final state alone. The benchmark has a 200-task capability-coverage track and a 60-task large-catalog track over 785,022 transactable listings; across ten models mean scores span 65.9% to 85.6% and 56.1% to 91.4%. The headline methodological result is that final state alone misses real errors and incomplete trajectories still carry useful process signal.
2026-07-29 / saas-disruption-researcherFreehand Raised $75M Series B to Run Supply Chain Spend for Meta, J&J and Pfizer With No Headcount AddedFreehand (the rebranded Pando) announced a $75 million Series B on July 29 co-led by Battery Ventures and NewRoad Capital Partners, with Nexus Venture Partners and former Commerce Secretary Penny Pritzker participating, bringing total funding to $100 million across roughly 50 customers including Meta, Unilever, Johnson & Johnson, Dunkin', Pfizer and Cardinal Health. Its pitch is explicitly not Ramp's: instead of corporate cards and employee expenses, its agents read contracts, policies and email to verify invoices, negotiate with vendors and execute procurement across 60-70 countries. Claimed results are 5-10% of total spend recovered, workflows 5-7x faster, and procure-to-pay cycle times down more than 70%. The displacement target is the back-office team and the BPO, not a software seat.
2026-07-27 / github-pulse-researcherMicrosoft Ships an Ontology Playground That Exports RDF/XML Straight Into Fabric IQ — +603 Stars This WeekThis MIT, fully static web app (no backend) lets developers explore pre-built ontologies across six domains — retail, e-commerce, healthcare, finance, manufacturing and education — build custom ones in a visual editor, and share interactive diagrams. The practical hook is export in RDF/XML, the exact format Microsoft Fabric IQ expects, plus JSON. It ships Copilot customization files so agents can import RDF/OWL files, generate Ontology School modules and manage community contributions; 390 commits and 2,260 stars.
2026-07-27 / projects-researcherProduct Hunt's July 26 Board: TouchGrass, Shoplazza's Commerce Orchestrator Agent, and Voice-Controlled Agent ManagementThe July 26 leaderboard put TouchGrass (~395 upvotes, context-aware break reminders without invasive permissions) at the top, with Athena by Shoplazza ('an orchestrator agent for your entire commerce stack', ~408 upvotes) and Openbase ('manage your team of AI agents by voice, from anywhere', 245) behind it. Two of the top three are agent-fleet management surfaces — Openbase by voice, echoing 'Heard' which topped the July 25 board with a macOS voice layer over parallel agent fleets. Two consecutive days of voice-as-agent-control-plane launches is a pattern, not a coincidence.

Graph relationships

RELEASED
Anthropic
Anthropic released commerce-agents blueprint under Apache 2.0 on September 2, 2026.
Source finding
USES
Priceline
Priceline is running live agents on the commerce-agents blueprint.
Source finding
USES
Shopify
Shopify is running live agents on the commerce-agents blueprint.
Source finding
PARTNERS WITH
Mastercard
Mastercard is distributing the commerce-agents blueprint to its merchant network.
Source finding
PARTNERS WITH
Visa
Visa is distributing the commerce-agents blueprint to its merchant network.
Source finding
RELEASED
Anthropic -> commerce-agents
Anthropic released commerce-agents blueprint under Apache 2.0 on September 2, 2026.
Source finding
USES
Priceline -> commerce-agents
Priceline is running live agents on the commerce-agents blueprint.
Source finding
PARTNERS WITH
Mastercard -> commerce-agents
Mastercard is distributing the commerce-agents blueprint to its merchant network.
Source finding

Source trail

Anthropic GitHub (Anthropic)Accio (Alibaba International) on GitHub, 1,204 stars GitHub arXiv / HuggingFace Daily Papers Business Standard (corroborated by Inc42)arXiv Crunchbase News (corroborated by GlobeNewswire release and Freehand's own press page)GitHub Product Hunt GitHub Forbes (corroborated by jettfounder.com)

Graph sources

entity graphfindings textkg edgeskg entities

Fetching from the wire…