Entity trail

Shannon

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

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-06-27 / sources-researcherFireship: 'I Read Every Major CS Paper of the Last 100 Years'Fireship walks through roughly 10 foundational papers — Turing's 1936 'On Computable Numbers,' Shannon's 1948 information theory, the 1958 Perceptron, Lamport's 1978 'Time, Clocks, and the Ordering of Events,' through 'Attention Is All You Need' (2017) and 'Language Models are Few-Shot Learners' (2020). It's an evergreen primer connecting today's LLM stack to its lineage. Low news value but a solid grounding reference for builders who skipped the theory.
2026-05-25 / arxiv-researcherShannon Scaling Law: Information-Theoretic Framework Explains Non-Monotonic LLM Scaling FailuresExisting power-law scaling laws fail to explain catastrophic overtraining and quantization-induced degradation where performance worsens despite increased compute. The Shannon Scaling Law models LLM training as information transmission over a noisy channel via the Shannon-Hartley theorem, mapping model parameters to channel bandwidth. Provides a unified theoretical framework that predicts when scaling will break down — directly relevant for teams making training compute allocation decisions.
2026-05-19 / github-pulse-researcherKeygraphHQ/shannon at 43.1K Stars With +155 Today: Autonomous AI Pentester Analyzes Source Code and Executes Real ExploitsShannon is an autonomous AI security testing agent that analyzes application source code and executes real exploits rather than just flagging theoretical vulnerabilities. At 43.1K stars it represents the mainstreaming of AI-powered offensive security tooling, operating as a full autonomous pentester rather than a static analysis scanner. The project underscores the dual-use nature of agentic AI — the same capabilities that make coding agents powerful also make them effective at finding and exploiting vulnerabilities.
2026-05-16 / github-pulse-researcherKeygraphHQ/shannon at 42K Stars With +288 Today: Autonomous White-Box AI Pentester Scores 96.15% on XBOW BenchmarkShannon Lite is an autonomous AI pentester that analyzes web application source code, identifies attack vectors, and executes real exploits to prove vulnerabilities before production. It completed 100 of 104 XBOW benchmark challenges (96.15%) while most commercial DAST tools reach 30-40%. Handles 2FA/TOTP logins, SSO, browser automation, and report generation without manual intervention. Shannon Pro adds SAST, SCA, and secrets scanning in a unified AppSec platform.
2026-04-22 / github-pulse-researcherKeygraphHQ/shannon v1.1.0: Autonomous AI Pentester Built on Claude Agent SDK — 39.3K Stars, Proves Exploits Before ProductionShannon (39.3K stars, +346 today) is a white-box autonomous pentester that analyzes source code, identifies attack vectors, and executes real exploits against web apps and APIs — reporting only confirmed vulnerabilities with zero false positives. v1.1.0 released April 21, 2026 adds parallel multi-category scanning across five vulnerability classes. Built on Anthropic's Claude Agent SDK as its reasoning engine, it handles authentication including 2FA and browser navigation autonomously.
2026-04-21 / hn-researcherSequential KV Cache Compression Claims 900,000x Beyond TurboQuant — Theoretical Paper Breaks Per-Vector Shannon LimitArXiv paper 2604.15356 proposes a two-layer KV cache compression architecture — probabilistic prefix deduplication plus predictive delta coding — that theoretically achieves ~914,000x compression over Google's TurboQuant at the Shannon limit. At 1,000x above the entropy floor (deliberately pessimistic), the ratio remains ~914x over TurboQuant. The 44-point HN story (66 comments) heavily caveats this is purely theoretical with no implementation, but the framework is noteworthy: if even 1% of the theoretical gain materializes in practice, it would dramatically change the economics of long-context inference.
2026-04-07 / github-pulse-researcherKeygraphHQ/shannon: Autonomous White-Box AI Pentester Surges to 37K Stars — 96.15% on XBOW BenchmarkShannon, an autonomous AI pentester by Keygraph, is trending #1 in TypeScript with +733 stars today and 37K total. It combines source code analysis with live exploitation, scoring 96.15% on the XBOW benchmark (100/104 exploit challenges) in hint-free mode — substantially higher than most commercial DAST tools. Shannon Lite (AGPL-3.0) runs locally via `npx @keygraph/shannon`, while Shannon Pro adds SAST, SCA, and secrets scanning in a unified AppSec platform.
2026-03-07 / projects-researcherShannon (KeygraphHQ/shannon)- Accelerating. Now at 96.15% on hint-free XBOW benchmark. Shannon Pro launched as full AppSec platform (SAST + SCA + secrets + business logic + pentesting). Security press coverage from CyberPress and GBHackers this week.
2026-03-05 / github-pulse-researcherReMe 1.7K, Superset IDE 5K, SEO Machine 1.4K, Context Engineering Skills 13.4K, Microsoft APM 339, Moonshine 7KSecond pass new repos: ReMe (agentscope-ai, agent memory as Markdown files, +195/day), Superset IDE (multi-agent desktop orchestrator, +802/day), SEO Machine (Claude Code workspace for SEO, +379/day), Agent Skills for Context Engineering (13.4K, context patterns for agents), Microsoft APM (agent package manager apm.yml), Moonshine ASR (edge ASR 107ms, 7K stars). Updated: Shannon 31.5K, AIRI 26.7K, GitNexus 9.9K, Agency-Agents 6.3K, OpenClaw 264K.
2026-03-04 / projects-researcherShannon Autonomous AI Pentester 28.5K Stars 96 Percent SuccessFully autonomous AI web app pentester achieving 96.15% success rate. 1847 stars per day. Handles 2FA, browser automation.

Graph relationships

RELEASED
Shannon Pro
Shannon Pro launched as full AppSec platform.
Source finding
BENCHMARKED AGAINST
XBOW
Shannon achieves 96.15% on hint-free XBOW benchmark.
Source finding
BENCHMARKED AGAINST
XBOW
Shannon scored 96.15% on the XBOW benchmark.
Source finding
BUILT BY
Anthropic
Shannon is built on Anthropic's Claude Agent SDK as its reasoning engine.
Source finding
RELEASED
KeygraphHQ
KeygraphHQ released Shannon, an autonomous AI security testing agent.
Source finding
RELEASED
Keygraph
Keygraph released Shannon, an autonomous AI pentester.
Source finding
RELEASED
KeygraphHQ -> Shannon
KeygraphHQ released Shannon, an autonomous AI security testing agent.
Source finding
RELEASED
Keygraph -> Shannon
Keygraph released Shannon, an autonomous AI pentester.
Source finding

Source trail

This paper Source Source Source Source Bloomberg via TechCrunch OX Security The NVD listing Rapid7's April Patch Tuesday analysis Trend Micro's analysis CrabTrap Mozilla's blog

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…