Skip to content

Intelligence Source Registry ​

This is the human-readable projection of the Digital Researcher's formal Watchlist. research/intelligence/REGISTRY.json remains the single source of truth; this page reads the Registry directly and does not maintain a second list.

People are the research subjects; repositories, papers, protocols, and official channels are the evidence and observation sources. Researcher counts and formal source-object counts are reported separately rather than added together.

Source Summary ​

47Core researchers
56Formal source objects
24P0 daily
26P1 weekly
6P2 focused
CategoryObjectsP0 DailyP1 WeeklyP2 FocusedNotes
AI Platforms660045 registered official channels
GitHub Engineering Repositories359206Core engineering, audit evaluation, work practice, and anomaly detection
Published Research15960Papers, research labs, artifacts, and curated research indexes

Core Researchers · 47 ​

Chi Wang

S+

Multi-agent framework author / researcher

multi-agent topologycoordination protocolslong-running harness
Related repositories

Qingyun Wu

S+

Academic researcher / Multi-Agent / framework author

multi-agent orchestrationagent networksevaluation
Related repositories

Charles Packer

S+

Agent memory / stateful runtime / founder

persistent identitylong-term memorystateful runtime
Related repositories

Sarah Wooders

S+

Agent memory / stateful runtime / research engineering

memory systemspersistent agentsruntime state
Related repositories

Omar Khattab

S

Academic researcher / AI systems / framework author

DSPyprogram optimizationagent systems
Related repositories

Shunyu Yao

S+

Researcher / reasoning and agents

reasoning agentssoftware agentslong-horizon tasks
Related repositories

Shuyan Zhou

S

Academic researcher / realistic agent environments

realistic environmentsweb agentsevaluation
Related repositories

Dawn Song

A+

AI safety and security researcher

agent securitytrustworthy AIdecentralized AI
Papers / protocols / projects

João Moura

S

Multi-agent framework author / founder

agent crewsrolesproduction orchestration
Related repositories

Swyx / Shawn Wang

A+

AI Engineer ecosystem / research media

agent engineeringdeveloper ecosystemindustry synthesis
Papers / protocols / projects

Tommaso Moro

A+

Agent / MCP engineering / GitHub

GitHub MCPagent skillsdeveloper workflows
Related repositories
Papers / protocols / projects

Robert Brennan

A+

Open-source agent platform founder / OpenHands

agent platformcoding runtimeproductization
Related repositories

Valerie Chen

S

Academic researcher / human-agent collaboration / verification

human-agent collaborationverificationevaluation
Related repositories
Papers / protocols / projects

Xuhui Zhou

S

Academic researcher / social and safe agents

social agentsagent safetyevaluation
Papers / protocols / projects

Aaron Parecki

A+

Identity protocol expert / OAuth / MCP Auth

OAuthMCP authorizationidentity
Papers / protocols / projects

Shreya Shankar

A+

Data systems / AI evals / academic research

AI evaluationdata systemsreliable pipelines
Papers / protocols / projects

Hamel Husain

A+

Engineering practitioner / AI evals / reliability

evalsagent reliabilityengineering practice
Related repositories
Papers / protocols / projects

Wilson Lin

S+

Cursor / long-running and multi-agent runtime

agent swarmslong-running agentsscaling agents
Papers / protocols / projects

Eric Traut

S

OpenAI Codex / runtime collaborator

agent loopsandboxpermissionssession / resume
Related repositories

David Gomes

S

Cursor / agent autonomy and auto-review

auto-reviewagent autonomycoding workflow
Papers / protocols / projects

Travis McPeak

S

Cursor / agent security and governance

agent securityhooksgovernance
Papers / protocols / projects

Ani Betts

S

Cursor / agent sandbox and MCP

sandboxMCPdesktop runtime
Papers / protocols / projects

Josh Ma

S

Cursor / cloud agent runtime

cloud agentsautomationslong-running execution
Papers / protocols / projects

Aman Sanger

A+

Cursor founder / agent product and architecture

agent productcloud codingruntime architecture
Papers / protocols / projects

AI Platforms · P0 Daily ​

OpenAI

ChatGPT · Codex · OpenAI API · Agents

ChatGPT, Codex, API and Agents product changes, documentation, enterprise controls and service status

P0daily11 public official channels
Official noticeAnnouncements ↗platform facts require official staff
View 5 precision sections

Precision sections produce research leads only; reproduction or official corroboration is required for a platform fact.

View 5 documentation, GitHub and status channels

Anthropic

Claude · Claude Code · Claude API

Claude, Claude Code and API product changes, agent runtime and governance

P0daily5 public official channels
Community not currently monitoredcommunity-discord requires authentication and is excluded from the daily due list.
View 5 documentation, GitHub and status channels

Google

Gemini · Gemini API · AI Studio · Vertex AI · Gemini CLI · ADK

Gemini, AI Studio, Vertex AI, Gemini CLI and ADK product and engineering changes

P0daily9 public official channels
Official noticeAnnouncement tag ↗official staff authorship must be verified
View 3 precision sections

Precision sections produce research leads only; reproduction or official corroboration is required for a platform fact.

View 5 documentation, GitHub and status channels

Anysphere

Cursor · Cursor Agents · Cloud Agents · Cursor CLI

AI coding agents, Cloud Agents, CLI and developer workflow changes

P0daily8 public official channels
Official noticeAnnouncements ↗platform facts require official staff
View 4 precision sections

Precision sections produce research leads only; reproduction or official corroboration is required for a platform fact.

View 3 documentation, GitHub and status channels

GitHub

GitHub Copilot · Coding Agent · Copilot CLI · Code Review

Copilot Coding Agent, CLI, Code Review and GitHub-native agent capabilities

P0daily6 public official channels
Official noticeCopilot News and Announcements ↗platform facts require official staff
View 1 precision sections

Precision sections produce research leads only; reproduction or official corroboration is required for a platform fact.

View 4 documentation, GitHub and status channels

Microsoft

Microsoft 365 Copilot · Copilot Studio · Agent Builder · Power Platform

Microsoft 365 Copilot, Copilot Studio, Agent Builder and enterprise agent platform

P0daily6 public official channels
Official noticeCopilot Studio Blog ↗platform facts require official staff
View 1 precision sections

Precision sections produce research leads only; reproduction or official corroboration is required for a platform fact.

View 4 documentation, GitHub and status channels

GitHub Engineering Watchlist ​

openai/codex

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · digital-employee
P0daily
Visit GitHub ↗

openai/openai-agents-python

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · digital-employee · industry-architecture
P0daily
Visit GitHub ↗

anthropics/claude-code

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · digital-employee · industry-architecture
P0daily
Visit GitHub ↗

google-gemini/gemini-cli

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · digital-employee
P0daily
Visit GitHub ↗

google/adk-python

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · industry-architecture · digital-employee
P0daily
Visit GitHub ↗

github/copilot-cli

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · digital-employee
P0daily
Visit GitHub ↗

modelcontextprotocol/python-sdk

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · industry-architecture
P0daily
Visit GitHub ↗

modelcontextprotocol/typescript-sdk

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · industry-architecture
P0daily
Visit GitHub ↗

a2aproject/A2A

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · industry-architecture
P0daily
Visit GitHub ↗

microsoft/autogen

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · digital-employee
P1weekly
Visit GitHub ↗

microsoft/semantic-kernel

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · industry-architecture
P1weekly
Visit GitHub ↗

langchain-ai/langgraph

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · digital-employee
P1weekly
Visit GitHub ↗

crewAIInc/crewAI

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · digital-employee
P1weekly
Visit GitHub ↗

OpenHands/OpenHands

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · digital-employee
P1weekly
Visit GitHub ↗

browser-use/browser-use

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: digital-employee · open-source-engineering
P1weekly
Visit GitHub ↗

SWE-bench/SWE-bench

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · digital-employee
P1weekly
Visit GitHub ↗

SWE-agent/SWE-agent

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · digital-employee
P1weekly
Visit GitHub ↗

multi-agent-systems-failure-taxonomy/MAST

multi-agent-failure-taxonomy-and-traces

Serves: open-source-engineering · digital-employee
P1weeklymulti-agent-failure-taxonomy-and-traces
Watch focus and special-study triggerfailure taxonomy · annotated traces · task verification failures · inter-agent misalignment · evaluation pipeline

A taxonomy revision, new annotated trace release, benchmark or evaluator change, or reproducible result that can test TMPA evidence reconstruction or CodeFlowMu verification.

Visit GitHub ↗

langfuse/langfuse

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · industry-architecture
P1weekly
Visit GitHub ↗

Arize-ai/phoenix

Agent runtimes, protocols, SDKs, tools, recovery, evaluation, observability, and engineering changes.

Serves: open-source-engineering · industry-architecture
P1weekly
Visit GitHub ↗

yzhao062/auditable

auditable-agent-runtime

Serves: digital-employee · open-source-engineering · industry-architecture
P1weeklyauditable-agent-runtime
Watch focus and special-study triggeradapters · decision record format · replay · rollback

A new adapter, replay/rollback mechanism, or decision-record schema with reproducible evidence.

Visit GitHub ↗

yzhao062/grade

failure-attribution-benchmark

Serves: open-source-engineering · digital-employee
P1weeklyfailure-attribution-benchmark
Watch focus and special-study triggerdatasets · graph representation · failure localization · cross-corpus transfer

A new dataset, graph representation, localization result, or cross-corpus transfer result.

Visit GitHub ↗

yzhao062/catchbench

agent-audit-lifecycle-benchmark

Serves: open-source-engineering · digital-employee
P1weeklyagent-audit-lifecycle-benchmark
Watch focus and special-study triggerPRE LIVE POST boards · datasets and pinned revisions · failure localization and detection · evaluation methods · known limitations and independent reproduction

The paper is released, a board or dataset changes materially, an artifact limitation is resolved, or an independently reproducible result becomes relevant to TMPA evidence reconstruction or CodeFlowMu verification.

Visit GitHub ↗

yzhao062/anywhere-agents

agent-work-practice

Serves: digital-employee · open-source-engineering
P1weeklyagent-work-practice
Watch focus and special-study triggerCodex and Claude configuration · skill routing · dual-agent review · safety hooks

A reusable configuration-governance or multi-agent review mechanism relevant to FCoP or CodeFlowMu.

Visit GitHub ↗

HeadyZhang/agent-audit

pre-deployment-agent-security

Serves: open-source-engineering · digital-employee
P2biweekly or release-triggeredpre-deployment-agent-security
Watch focus and special-study triggerMCP configuration scanning · prompt injection · taint propagation · excessive permissions

A release that adds a reproducible scanner, rule set, or risk model applicable to deployment gates.

Visit GitHub ↗

Justin0504/Aegis

runtime-policy-enforcement

Serves: digital-employee · open-source-engineering · industry-architecture
P2biweekly or release-triggeredruntime-policy-enforcement
Watch focus and special-study triggerpre-tool policy checks · human approval · encrypted audit trail · runtime enforcement

A release that materially changes tool-call enforcement, approval, or audit-trail evidence.

Visit GitHub ↗

yzhao062/agent-style

agent-writing-quality

Serves: digital-employee · open-source-engineering
P2biweekly or release-triggeredagent-writing-quality
Watch focus and special-study triggerwriting rules · citation discipline · automated checks

A rule or checker that can be compared with the JoinWell52 editorial scorecard.

Visit GitHub ↗

yzhao062/cs-paper-checklist

research-publication-quality

Serves: open-source-engineering · digital-employee
P2biweekly or release-triggeredresearch-publication-quality
Watch focus and special-study triggerpaper completeness · method reporting · evaluation reporting · submission checks

A checklist change that can strengthen the TMPA publication gate without weakening evidence requirements.

Visit GitHub ↗

yzhao062/pyod

anomaly-detection-method

Serves: open-source-engineering
P2monthlyanomaly-detection-method
Watch focus and special-study triggernew detectors · evaluation changes · behavior anomaly applications

A method or evaluation change directly applicable to Agent runtime anomaly signals.

Visit GitHub ↗

yzhao062/anomaly-detection-resources

anomaly-detection-research-index

Serves: open-source-engineering · digital-employee
P2monthlyanomaly-detection-research-index
Watch focus and special-study triggernew surveys · benchmarks · tool collections · agent behavior anomaly work

A new cluster of resources that materially connects anomaly detection to Agent behavior or runtime governance.

Visit GitHub ↗

Kyvvu/manifests

runtime-path-policy

Serves: open-source-engineering · digital-employee · industry-architecture
P1weeklyruntime-path-policy
Watch focus and special-study triggerpath-dependent policy · pre-action decisions · policy tests · policy schema

Visit GitHub ↗

schroederdewitt/perfectly-secure-steganography

covert-agent-communication

Serves: open-source-engineering · digital-employee · industry-architecture
P1weeklycovert-agent-communication
Watch focus and special-study triggerminimum entropy coupling · steganography evaluation · observability limits

Visit GitHub ↗

mansura-habiba/bulbasaur-skill-cli

agent-skill-audit

Serves: open-source-engineering · digital-employee · industry-architecture
P1weeklyagent-skill-audit
Watch focus and special-study triggerskill contracts · trust audit · quarantine and review · evaluation · locked dependencies

Visit GitHub ↗

ag2ai/Agents_Failure_Attribution

agent-failure-attribution

Serves: open-source-engineering · digital-employee · industry-architecture
P1weeklyagent-failure-attribution
Watch focus and special-study triggerWho and When dataset · failure attribution · critical steps · evaluation scripts

Visit GitHub ↗

haoyuwang99/AgentSpec

agent-runtime-enforcement

Serves: open-source-engineering · digital-employee · industry-architecture
P1weeklyagent-runtime-enforcement
Watch focus and special-study triggerpolicy DSL · triggers and predicates · enforcement actions · evaluation and regressions

Visit GitHub ↗

Published Research ​

arXiv

Academic preprints; focused on cs.MA, cs.SE and cs.AI

cs.MA · cs.SE · cs.AI
P0daily
Visit source ↗

Autonomous Agents and Multi-Agent Systems

Autonomous Agents and Multi-Agent Systems; agent organization, coordination, protocols and governance

multi-agent systems · agent organization · coordination · protocol · governance
P0daily
Visit source ↗

Journal of Systems and Software

Journal of Systems and Software; architecture, engineering, empirical evaluation and reproducibility

software architecture · software engineering · empirical evaluation · reproducibility · agent engineering
P0daily
Visit source ↗

Zenodo

Open research repository; software, datasets, artifacts and reproducibility evidence

multi-agent · agent runtime · orchestration · software artifact · reproducibility
P0daily
Visit source ↗

OpenReview

Open peer-review platform

conference-paper · review
P0daily
Visit source ↗

OpenAI Research

Official OpenAI research

official-research · system-card · technical-report
P0daily
Visit source ↗

Anthropic Research

Official Anthropic research

official-research · system-card · technical-report
P0daily
Visit source ↗

Google DeepMind Research

Official Google DeepMind research

official-research · technical-report
P0daily
Visit source ↗

Microsoft Research

Microsoft Research; agents, enterprise AI, software engineering, systems and benchmarks

official-research · technical-report · benchmark
P0daily
Visit source ↗

ACL Anthology

ACL paper archive

conference-paper · journal-paper
P1weekly
Visit source ↗

ACM Digital Library

ACM Digital Library

conference-paper · journal-paper
P1weekly
Visit source ↗

IEEE Xplore

IEEE papers and standards

conference-paper · journal-paper · standard
P1weekly
Visit source ↗

Awesome Auditable AI

A curated index of papers, datasets, benchmarks, standards, and tools for auditable AI. It is a discovery source; every useful entry must be verified against its primary source.

agent auditability · agent reliability · failure attribution · runtime monitoring · decision records · standards and governance
P1weeklySecondary navigation source

Evidence rule: Follow every useful entry to the original paper, official repository, dataset, benchmark, or standard before using it as evidence.

Special-study triggerA major taxonomy change, new standard, high-value paper cluster, or mechanism directly relevant to TMPA, FCoP, or CodeFlowMu.
Visit source ↗

Notes ​

  • P0 / daily sources are formal daily Discovery inputs.
  • P1 / weekly sources broaden high-value coverage.
  • P2 / focused objects are checked biweekly, on releases, or monthly; the Registry also records their watch focus and special-study trigger.
  • Curated research indexes discover leads and never replace the original paper, official repository, dataset, benchmark, or standard.
  • Actual daily scan outcomes live in that day's Intelligence Run. Add or remove sources only in the Registry.

Last updated: