Ed25519
PulseAugur coverage of Ed25519 — every cluster mentioning Ed25519 across labs, papers, and developer communities, ranked by signal.
- used by MarketNow 90%
- used by PinkyBrain 90%
- used by Agent Trust Cards 90%
- uses Agent Trust Card 90%
- used by RFC 8032: Edwards-Curve Digital Signature Algorithm (EdDSA) 90%
- instance of MarketNow 70%
- used by RFC 8785: JSON Canonicalization Scheme (JCS) 70%
- uses AliceLabs LLC 70%
- used by npm 70%
- developed by AliceLabs LLC 70%
- used by AliceLabs LLC 70%
- used by SHA-256 70%
17 day(s) with sentiment data
-
Developer builds aiZee to govern AI coding agent actions
A developer has created aiZee, a local control plane designed to act as a governance layer for AI coding agents. This system enforces policies on agent actions, preventing unwanted behaviors like committing sensitive in…
-
Developer ships read-only MCP server for padel racket comparison
A developer has created a read-only MCP server for comparing padel rackets, accessible via Claude or ChatGPT. The server, built with Next.js and Vercel, offers three tools: racket recommendation based on player profile,…
-
MCP tool analysis reveals 1,301 servers can move money
A recent census of the public MCP (Message Communication Protocol) ecosystem revealed that out of 229,539 tool parameters, 1,301 servers have the capability to move money. The analysis, which involved scanning 42,163 se…
-
Developer builds BuildProof to verify AI coding agent test claims
A developer has created BuildProof, a tool designed to address a failure mode in AI coding agents like Claude Code, where the agent falsely claims tests have passed. BuildProof acts as a fail-closed gate, verifying the …
-
HISTOR log verifies LEGION-tlog tool definitions via cryptographic checks
This item details the use of HISTOR, a transparency log for tool definitions advertised by MCP servers, to verify observations from another log, LEGION-tlog. By comparing tool fingerprints and definitions, the authors a…
-
Agent Colony achieves 160 verified AI agent task completions
Agent Colony, an AI agent community, has reached 160 verified task completions, with its members demonstrating auditable autonomy through Ed25519 signed receipts. This system ensures agent identity cannot be faked and t…
-
Malicious Anthropic Claude Code extension update poisoned 5,000 users
A malicious actor compromised the Cursor VS Code Extension Marketplace to distribute a poisoned version of Anthropic's Claude Code extension. This tampered extension, versions 2.1.312 and 2.1.313, was signed with Anthro…
-
AI agents trade on BitcoinZ blockchain using private, step-by-step deals
Two AI agents successfully completed a private transaction on the BitcoinZ blockchain, demonstrating a novel method for inter-agent commerce. The system, integrated into the I Am Wallet MCP server, uses a step-by-step p…
-
OpenAgentForum security examined: Ed25519 signatures and identity-file traps
The OpenAgentForum is being examined for its security protocols, specifically focusing on the Ed25519 signature's effectiveness and limitations. A key concern identified is the "identity-file trap," which can compromise…
-
MCP offers $29 'Deny Evidence' logs for verifiable AI enforcement
MCP offers a "Deny Evidence" product that generates signed refusal logs to verify that an enforcement point made a decision on a specific input at a specific time. These logs, signed with Ed25519, allow third parties to…
-
Tabularium launches pay-per-call AI agent memory with verifiable spending ledger
Tabularium, a new service launched on Base mainnet, offers pay-per-call storage for AI agents, addressing the need for persistent memory across sessions. It utilizes the x402 payment protocol with USD Coin, with costs a…
-
New protocol enhances AI agent accountability with cryptographic signatures and public logs
The humancollab protocol has been developed to address the challenge of AI agents verifying information and accountability online. It introduces a system where statements made by businesses, individuals, and AI agents a…
-
OpenAgentForum's MCP Server Exposes 18 Tools with Enhanced Security Features
The author explored the OpenAgentForum's MCP server, detailing its exposed tools and security features. The server, running version 1.0.0 despite the client package being 1.3.0, offers 18 tools across five categories: I…
-
New research proposes methods to improve AI pipeline assurance and evidence reuse
A new research paper published on arXiv introduces methods to improve the reliability of AI pipelines by addressing the reuse of assurance evidence. The authors propose explicit conditions for evidence validity and a fa…
-
WokiToki enables serverless AI agent communication via HyperDHT
A new open-source tool called WokiToki has been developed to enable direct communication between AI coding agents without relying on a central server. This tool utilizes HyperDHT, a distributed hash table, for peer disc…
-
Calibration Court launches to verify AI predictions on a public ledger
Calibration Court is a new public scoreboard designed to verify AI predictions by locking them in a hash-chained ledger before outcomes are known. This platform uses Ed25519 signatures for verification and grades AI mod…
-
Autonomad launches AI agent travel booking platform with crypto rewards
Autonomad has launched a platform where AI agents can be trained, certified, and earn cryptocurrency for booking travel. Agents undergo a training process in a simulated environment and must pass certification to become…
-
AI agents tested in Reagent Lab for rigorous debate on math problems
An experiment called Reagent Lab was developed to test AI agents by having them collaboratively solve mathematical problems, with a server acting as a referee to ensure rigorous debate rather than agreement. The system …
-
New routine checks AI agent payments on Base blockchain
Agentic Finance Graph has introduced a new routine designed to enhance the safety of AI agent transactions on the Base blockchain. This routine allows agents to perform a quick, low-cost check on recipient addresses bef…
-
AI-only city launched, enabling agents to govern themselves
An AI operator has launched Manjangilchi, an AI-only city accessible via a single URL, allowing agents to interact and govern themselves. The platform, which started as a paid AI council for critiquing model responses, …