{ "generatedAt": "2026-05-25T07:18:51.368Z", "articleCount": 42, "editionCount": 33, "evidenceCount": 42, "atlasCount": 4, "latestEdition": { "date": "2026-05-25", "dateLabel": "May 25, 2026", "url": "/editions/2026-05-25/", "storyCount": 1 }, "editions": [ { "date": "2026-05-25", "dateLabel": "May 25, 2026", "url": "/editions/2026-05-25/", "storyCount": 1, "sourceCount": 5, "lead": { "title": "Agent Reliability Is Moving Into the Test Rig", "url": "/articles/2026-05-25-daily-agent-reliability-is-moving-into-the-test-rig/", "evidenceUrl": "/evidence/2026-05-25-daily-agent-reliability-is-moving-into-the-test-rig/" }, "articles": [ { "title": "Agent Reliability Is Moving Into the Test Rig", "url": "/articles/2026-05-25-daily-agent-reliability-is-moving-into-the-test-rig/", "evidenceUrl": "/evidence/2026-05-25-daily-agent-reliability-is-moving-into-the-test-rig/", "lane": "Daily Edition" } ] }, { "date": "2026-05-24", "dateLabel": "May 24, 2026", "url": "/editions/2026-05-24/", "storyCount": 1, "sourceCount": 6, "lead": { "title": "Agent Runtimes Are Learning Where to Say No", "url": "/articles/2026-05-24-daily-agent-runtimes-are-learning-where-to-say-no/", "evidenceUrl": "/evidence/2026-05-24-daily-agent-runtimes-are-learning-where-to-say-no/" }, "articles": [ { "title": "Agent Runtimes Are Learning Where to Say No", "url": "/articles/2026-05-24-daily-agent-runtimes-are-learning-where-to-say-no/", "evidenceUrl": "/evidence/2026-05-24-daily-agent-runtimes-are-learning-where-to-say-no/", "lane": "Daily Edition" } ] }, { "date": "2026-05-23", "dateLabel": "May 23, 2026", "url": "/editions/2026-05-23/", "storyCount": 1, "sourceCount": 13, "lead": { "title": "Agent Runtimes Are Learning to Ship Their Control Plane", "url": "/articles/2026-05-23-daily-agent-runtimes-are-learning-to-ship-their-control-plane/", "evidenceUrl": "/evidence/2026-05-23-daily-agent-runtimes-are-learning-to-ship-their-control-plane/" }, "articles": [ { "title": "Agent Runtimes Are Learning to Ship Their Control Plane", "url": "/articles/2026-05-23-daily-agent-runtimes-are-learning-to-ship-their-control-plane/", "evidenceUrl": "/evidence/2026-05-23-daily-agent-runtimes-are-learning-to-ship-their-control-plane/", "lane": "Daily Edition" } ] }, { "date": "2026-05-22", "dateLabel": "May 22, 2026", "url": "/editions/2026-05-22/", "storyCount": 1, "sourceCount": 8, "lead": { "title": "Agent Runtimes Are Making Their Limits Explicit", "url": "/articles/2026-05-22-daily-agent-runtimes-are-making-their-limits-explicit/", "evidenceUrl": "/evidence/2026-05-22-daily-agent-runtimes-are-making-their-limits-explicit/" }, "articles": [ { "title": "Agent Runtimes Are Making Their Limits Explicit", "url": "/articles/2026-05-22-daily-agent-runtimes-are-making-their-limits-explicit/", "evidenceUrl": "/evidence/2026-05-22-daily-agent-runtimes-are-making-their-limits-explicit/", "lane": "Daily Edition" } ] }, { "date": "2026-05-21", "dateLabel": "May 21, 2026", "url": "/editions/2026-05-21/", "storyCount": 1, "sourceCount": 7, "lead": { "title": "Agent Goals Are Becoming Runtime State", "url": "/articles/2026-05-21-daily-agent-goals-are-becoming-runtime-state/", "evidenceUrl": "/evidence/2026-05-21-daily-agent-goals-are-becoming-runtime-state/" }, "articles": [ { "title": "Agent Goals Are Becoming Runtime State", "url": "/articles/2026-05-21-daily-agent-goals-are-becoming-runtime-state/", "evidenceUrl": "/evidence/2026-05-21-daily-agent-goals-are-becoming-runtime-state/", "lane": "Daily Edition" } ] }, { "date": "2026-05-20", "dateLabel": "May 20, 2026", "url": "/editions/2026-05-20/", "storyCount": 1, "sourceCount": 9, "lead": { "title": "Agent Runtimes Are Moving State Out of the Shadows", "url": "/articles/2026-05-20-daily-agent-runtimes-are-moving-state-out-of-the-shadows/", "evidenceUrl": "/evidence/2026-05-20-daily-agent-runtimes-are-moving-state-out-of-the-shadows/" }, "articles": [ { "title": "Agent Runtimes Are Moving State Out of the Shadows", "url": "/articles/2026-05-20-daily-agent-runtimes-are-moving-state-out-of-the-shadows/", "evidenceUrl": "/evidence/2026-05-20-daily-agent-runtimes-are-moving-state-out-of-the-shadows/", "lane": "Daily Edition" } ] }, { "date": "2026-05-19", "dateLabel": "May 19, 2026", "url": "/editions/2026-05-19/", "storyCount": 1, "sourceCount": 8, "lead": { "title": "Agent Runtimes Are Learning to Audit Their Own Tools", "url": "/articles/2026-05-19-daily-agent-runtimes-are-learning-to-audit-their-own-tools/", "evidenceUrl": "/evidence/2026-05-19-daily-agent-runtimes-are-learning-to-audit-their-own-tools/" }, "articles": [ { "title": "Agent Runtimes Are Learning to Audit Their Own Tools", "url": "/articles/2026-05-19-daily-agent-runtimes-are-learning-to-audit-their-own-tools/", "evidenceUrl": "/evidence/2026-05-19-daily-agent-runtimes-are-learning-to-audit-their-own-tools/", "lane": "Daily Edition" } ] }, { "date": "2026-04-01", "dateLabel": "April 1, 2026", "url": "/editions/2026-04-01/", "storyCount": 2, "sourceCount": 6, "lead": { "title": "Agent CLIs Are Turning Permissions Into a Conversation", "url": "/articles/2026-04-01-daily-agent-clis-are-turning-permissions-into-a-conversation/", "evidenceUrl": "/evidence/2026-04-01-daily-agent-clis-are-turning-permissions-into-a-conversation/" }, "articles": [ { "title": "Agent CLIs Are Turning Permissions Into a Conversation", "url": "/articles/2026-04-01-daily-agent-clis-are-turning-permissions-into-a-conversation/", "evidenceUrl": "/evidence/2026-04-01-daily-agent-clis-are-turning-permissions-into-a-conversation/", "lane": "Daily Edition" }, { "title": "AI Agents Are Turning Compaction Into State Surgery", "url": "/articles/2026-04-01-agent-daily-ai-agents-are-turning-compaction-into-state-surgery/", "evidenceUrl": "/evidence/2026-04-01-agent-daily-ai-agents-are-turning-compaction-into-state-surgery/", "lane": "Agent Daily" } ] }, { "date": "2026-03-31", "dateLabel": "March 31, 2026", "url": "/editions/2026-03-31/", "storyCount": 2, "sourceCount": 2, "lead": { "title": "The Real Agent Feature Is Not Losing the Plot", "url": "/articles/2026-03-31-daily-the-real-agent-feature-is-not-losing-the-plot/", "evidenceUrl": "/evidence/2026-03-31-daily-the-real-agent-feature-is-not-losing-the-plot/" }, "articles": [ { "title": "The Real Agent Feature Is Not Losing the Plot", "url": "/articles/2026-03-31-daily-the-real-agent-feature-is-not-losing-the-plot/", "evidenceUrl": "/evidence/2026-03-31-daily-the-real-agent-feature-is-not-losing-the-plot/", "lane": "Daily Edition" }, { "title": "AI Agent Summaries Are Becoming Infrastructure", "url": "/articles/2026-03-31-agent-daily-ai-agent-summaries-are-becoming-infrastructure/", "evidenceUrl": "/evidence/2026-03-31-agent-daily-ai-agent-summaries-are-becoming-infrastructure/", "lane": "Agent Daily" } ] }, { "date": "2026-03-30", "dateLabel": "March 30, 2026", "url": "/editions/2026-03-30/", "storyCount": 2, "sourceCount": 8, "lead": { "title": "The CLI Is Becoming an Agent Workbench", "url": "/articles/2026-03-30-daily-the-cli-is-becoming-an-agent-workbench/", "evidenceUrl": "/evidence/2026-03-30-daily-the-cli-is-becoming-an-agent-workbench/" }, "articles": [ { "title": "The CLI Is Becoming an Agent Workbench", "url": "/articles/2026-03-30-daily-the-cli-is-becoming-an-agent-workbench/", "evidenceUrl": "/evidence/2026-03-30-daily-the-cli-is-becoming-an-agent-workbench/", "lane": "Daily Edition" }, { "title": "The Next Agent Battle Isn’t Subagents. It’s Delegation Quality.", "url": "/articles/2026-03-30-agent-daily-the-next-agent-battle-isn-t-subagents-it-s-delegation-quality/", "evidenceUrl": "/evidence/2026-03-30-agent-daily-the-next-agent-battle-isn-t-subagents-it-s-delegation-quality/", "lane": "Agent Daily" } ] }, { "date": "2026-03-29", "dateLabel": "March 29, 2026", "url": "/editions/2026-03-29/", "storyCount": 2, "sourceCount": 16, "lead": { "title": "From Tool Chatter to Chapters: Agent CLIs Are Inventing a Narrative Layer", "url": "/articles/2026-03-29-daily-from-tool-chatter-to-chapters-agent-clis-are-inventing-a-narrative-layer/", "evidenceUrl": "/evidence/2026-03-29-daily-from-tool-chatter-to-chapters-agent-clis-are-inventing-a-narrative-layer/" }, "articles": [ { "title": "From Tool Chatter to Chapters: Agent CLIs Are Inventing a Narrative Layer", "url": "/articles/2026-03-29-daily-from-tool-chatter-to-chapters-agent-clis-are-inventing-a-narrative-layer/", "evidenceUrl": "/evidence/2026-03-29-daily-from-tool-chatter-to-chapters-agent-clis-are-inventing-a-narrative-layer/", "lane": "Daily Edition" }, { "title": "AI Coding Agents Are Turning Compaction Into a Policy Engine", "url": "/articles/2026-03-29-agent-daily-ai-coding-agents-are-turning-compaction-into-a-policy-engine/", "evidenceUrl": "/evidence/2026-03-29-agent-daily-ai-coding-agents-are-turning-compaction-into-a-policy-engine/", "lane": "Agent Daily" } ] }, { "date": "2026-03-28", "dateLabel": "March 28, 2026", "url": "/editions/2026-03-28/", "storyCount": 2, "sourceCount": 12, "lead": { "title": "The Next Agent UX Moat Isn’t Speed. It’s Backpressure.", "url": "/articles/2026-03-28-daily-the-next-agent-ux-moat-isn-t-speed-it-s-backpressure/", "evidenceUrl": "/evidence/2026-03-28-daily-the-next-agent-ux-moat-isn-t-speed-it-s-backpressure/" }, "articles": [ { "title": "The Next Agent UX Moat Isn’t Speed. It’s Backpressure.", "url": "/articles/2026-03-28-daily-the-next-agent-ux-moat-isn-t-speed-it-s-backpressure/", "evidenceUrl": "/evidence/2026-03-28-daily-the-next-agent-ux-moat-isn-t-speed-it-s-backpressure/", "lane": "Daily Edition" }, { "title": "AI Coding Agents Have a New Bottleneck: Keeping Work Intact", "url": "/articles/2026-03-28-agent-daily-ai-coding-agents-have-a-new-bottleneck-keeping-work-intact/", "evidenceUrl": "/evidence/2026-03-28-agent-daily-ai-coding-agents-have-a-new-bottleneck-keeping-work-intact/", "lane": "Agent Daily" } ] }, { "date": "2026-03-27", "dateLabel": "March 27, 2026", "url": "/editions/2026-03-27/", "storyCount": 2, "sourceCount": 12, "lead": { "title": "Subagents Aren’t Just Getting Smarter. They’re Getting Contained.", "url": "/articles/2026-03-27-daily-subagents-aren-t-just-getting-smarter-they-re-getting-contained/", "evidenceUrl": "/evidence/2026-03-27-daily-subagents-aren-t-just-getting-smarter-they-re-getting-contained/" }, "articles": [ { "title": "Subagents Aren’t Just Getting Smarter. They’re Getting Contained.", "url": "/articles/2026-03-27-daily-subagents-aren-t-just-getting-smarter-they-re-getting-contained/", "evidenceUrl": "/evidence/2026-03-27-daily-subagents-aren-t-just-getting-smarter-they-re-getting-contained/", "lane": "Daily Edition" }, { "title": "AI Coding Agents Are Turning Approval Settings Into Operating Modes", "url": "/articles/2026-03-27-agent-daily-ai-coding-agents-are-turning-approval-settings-into-operating-modes/", "evidenceUrl": "/evidence/2026-03-27-agent-daily-ai-coding-agents-are-turning-approval-settings-into-operating-modes/", "lane": "Agent Daily" } ] }, { "date": "2026-03-26", "dateLabel": "March 26, 2026", "url": "/editions/2026-03-26/", "storyCount": 2, "sourceCount": 8, "lead": { "title": "The CLI Is Quietly Becoming an Agent Router", "url": "/articles/2026-03-26-daily-the-cli-is-quietly-becoming-an-agent-router/", "evidenceUrl": "/evidence/2026-03-26-daily-the-cli-is-quietly-becoming-an-agent-router/" }, "articles": [ { "title": "The CLI Is Quietly Becoming an Agent Router", "url": "/articles/2026-03-26-daily-the-cli-is-quietly-becoming-an-agent-router/", "evidenceUrl": "/evidence/2026-03-26-daily-the-cli-is-quietly-becoming-an-agent-router/", "lane": "Daily Edition" }, { "title": "Web Fetch Is Emerging as a Security Boundary for AI Agents", "url": "/articles/2026-03-26-agent-daily-web-fetch-is-emerging-as-a-security-boundary-for-ai-agents/", "evidenceUrl": "/evidence/2026-03-26-agent-daily-web-fetch-is-emerging-as-a-security-boundary-for-ai-agents/", "lane": "Agent Daily" } ] }, { "date": "2026-03-25", "dateLabel": "March 25, 2026", "url": "/editions/2026-03-25/", "storyCount": 2, "sourceCount": 5, "lead": { "title": "The New CLI Moat Isn’t UX. It’s How Agent Skills Get Shipped", "url": "/articles/2026-03-25-daily-the-new-cli-moat-isn-t-ux-it-s-how-agent-skills-get-shipped/", "evidenceUrl": "/evidence/2026-03-25-daily-the-new-cli-moat-isn-t-ux-it-s-how-agent-skills-get-shipped/" }, "articles": [ { "title": "The New CLI Moat Isn’t UX. It’s How Agent Skills Get Shipped", "url": "/articles/2026-03-25-daily-the-new-cli-moat-isn-t-ux-it-s-how-agent-skills-get-shipped/", "evidenceUrl": "/evidence/2026-03-25-daily-the-new-cli-moat-isn-t-ux-it-s-how-agent-skills-get-shipped/", "lane": "Daily Edition" }, { "title": "Sandboxing Is Starting to Look Like a Runtime Layer for AI Coding Agents", "url": "/articles/2026-03-25-agent-daily-sandboxing-is-starting-to-look-like-a-runtime-layer-for-ai-coding-agents/", "evidenceUrl": "/evidence/2026-03-25-agent-daily-sandboxing-is-starting-to-look-like-a-runtime-layer-for-ai-coding-agents/", "lane": "Agent Daily" } ] }, { "date": "2026-03-24", "dateLabel": "March 24, 2026", "url": "/editions/2026-03-24/", "storyCount": 2, "sourceCount": 6, "lead": { "title": "The Next CLI UX Battle Is Agent Forensics", "url": "/articles/2026-03-24-daily-the-next-cli-ux-battle-is-agent-forensics/", "evidenceUrl": "/evidence/2026-03-24-daily-the-next-cli-ux-battle-is-agent-forensics/" }, "articles": [ { "title": "The Next CLI UX Battle Is Agent Forensics", "url": "/articles/2026-03-24-daily-the-next-cli-ux-battle-is-agent-forensics/", "evidenceUrl": "/evidence/2026-03-24-daily-the-next-cli-ux-battle-is-agent-forensics/", "lane": "Daily Edition" }, { "title": "Subagents Are Getting Job Titles, Badge Checks, and a Manager Chain", "url": "/articles/2026-03-24-agent-daily-subagents-are-getting-job-titles-badge-checks-and-a-manager-chain/", "evidenceUrl": "/evidence/2026-03-24-agent-daily-subagents-are-getting-job-titles-badge-checks-and-a-manager-chain/", "lane": "Agent Daily" } ] }, { "date": "2026-03-23", "dateLabel": "March 23, 2026", "url": "/editions/2026-03-23/", "storyCount": 1, "sourceCount": 3, "lead": { "title": "Codex forks it, Gemini threads it: execution context becomes first-class", "url": "/articles/2026-03-23-daily-codex-forks-it-gemini-threads-it-execution-context-becomes-first-class/", "evidenceUrl": "/evidence/2026-03-23-daily-codex-forks-it-gemini-threads-it-execution-context-becomes-first-class/" }, "articles": [ { "title": "Codex forks it, Gemini threads it: execution context becomes first-class", "url": "/articles/2026-03-23-daily-codex-forks-it-gemini-threads-it-execution-context-becomes-first-class/", "evidenceUrl": "/evidence/2026-03-23-daily-codex-forks-it-gemini-threads-it-execution-context-becomes-first-class/", "lane": "Daily Edition" } ] }, { "date": "2026-03-22", "dateLabel": "March 22, 2026", "url": "/editions/2026-03-22/", "storyCount": 1, "sourceCount": 4, "lead": { "title": "Parallel agents are getting real addresses", "url": "/articles/2026-03-22-daily-parallel-agents-are-getting-real-addresses/", "evidenceUrl": "/evidence/2026-03-22-daily-parallel-agents-are-getting-real-addresses/" }, "articles": [ { "title": "Parallel agents are getting real addresses", "url": "/articles/2026-03-22-daily-parallel-agents-are-getting-real-addresses/", "evidenceUrl": "/evidence/2026-03-22-daily-parallel-agents-are-getting-real-addresses/", "lane": "Daily Edition" } ] }, { "date": "2026-03-21", "dateLabel": "March 21, 2026", "url": "/editions/2026-03-21/", "storyCount": 1, "sourceCount": 1, "lead": { "title": "AI agents are getting better at saying “here’s what I finished”", "url": "/articles/2026-03-21-daily-ai-agents-are-getting-better-at-saying-here-s-what-i-finished/", "evidenceUrl": "/evidence/2026-03-21-daily-ai-agents-are-getting-better-at-saying-here-s-what-i-finished/" }, "articles": [ { "title": "AI agents are getting better at saying “here’s what I finished”", "url": "/articles/2026-03-21-daily-ai-agents-are-getting-better-at-saying-here-s-what-i-finished/", "evidenceUrl": "/evidence/2026-03-21-daily-ai-agents-are-getting-better-at-saying-here-s-what-i-finished/", "lane": "Daily Edition" } ] }, { "date": "2026-03-20", "dateLabel": "March 20, 2026", "url": "/editions/2026-03-20/", "storyCount": 1, "sourceCount": 2, "lead": { "title": "Before the Prompt Lands: Codex and Gemini Turn Hooks Into Agent Control Planes", "url": "/articles/2026-03-20-daily-before-the-prompt-lands-codex-and-gemini-turn-hooks-into-agent-control-planes/", "evidenceUrl": "/evidence/2026-03-20-daily-before-the-prompt-lands-codex-and-gemini-turn-hooks-into-agent-control-planes/" }, "articles": [ { "title": "Before the Prompt Lands: Codex and Gemini Turn Hooks Into Agent Control Planes", "url": "/articles/2026-03-20-daily-before-the-prompt-lands-codex-and-gemini-turn-hooks-into-agent-control-planes/", "evidenceUrl": "/evidence/2026-03-20-daily-before-the-prompt-lands-codex-and-gemini-turn-hooks-into-agent-control-planes/", "lane": "Daily Edition" } ] }, { "date": "2026-03-19", "dateLabel": "March 19, 2026", "url": "/editions/2026-03-19/", "storyCount": 1, "sourceCount": 5, "lead": { "title": "Gemini Lets the Model Schedule Parallel Tools. Codex Makes the Runtime Decide.", "url": "/articles/2026-03-19-daily-gemini-lets-the-model-schedule-parallel-tools-codex-makes-the-runtime-decide/", "evidenceUrl": "/evidence/2026-03-19-daily-gemini-lets-the-model-schedule-parallel-tools-codex-makes-the-runtime-decide/" }, "articles": [ { "title": "Gemini Lets the Model Schedule Parallel Tools. Codex Makes the Runtime Decide.", "url": "/articles/2026-03-19-daily-gemini-lets-the-model-schedule-parallel-tools-codex-makes-the-runtime-decide/", "evidenceUrl": "/evidence/2026-03-19-daily-gemini-lets-the-model-schedule-parallel-tools-codex-makes-the-runtime-decide/", "lane": "Daily Edition" } ] }, { "date": "2026-03-18", "dateLabel": "March 18, 2026", "url": "/editions/2026-03-18/", "storyCount": 1, "sourceCount": 6, "lead": { "title": "Subagents Grow Up: Gemini Isolates Tool Boundaries While Codex Shares Trust by Default", "url": "/articles/2026-03-18-daily-subagents-grow-up-gemini-isolates-tool-boundaries-while-codex-shares-trust-by-default/", "evidenceUrl": "/evidence/2026-03-18-daily-subagents-grow-up-gemini-isolates-tool-boundaries-while-codex-shares-trust-by-default/" }, "articles": [ { "title": "Subagents Grow Up: Gemini Isolates Tool Boundaries While Codex Shares Trust by Default", "url": "/articles/2026-03-18-daily-subagents-grow-up-gemini-isolates-tool-boundaries-while-codex-shares-trust-by-default/", "evidenceUrl": "/evidence/2026-03-18-daily-subagents-grow-up-gemini-isolates-tool-boundaries-while-codex-shares-trust-by-default/", "lane": "Daily Edition" } ] }, { "date": "2026-03-15", "dateLabel": "March 15, 2026", "url": "/editions/2026-03-15/", "storyCount": 1, "sourceCount": 1, "lead": { "title": "OpenViking Turns Agent Memory into a Filesystem — and That Changes the Game", "url": "/articles/2026-03-15-daily-openviking-turns-agent-memory-into-a-filesystem-and-that-changes-the-game/", "evidenceUrl": "/evidence/2026-03-15-daily-openviking-turns-agent-memory-into-a-filesystem-and-that-changes-the-game/" }, "articles": [ { "title": "OpenViking Turns Agent Memory into a Filesystem — and That Changes the Game", "url": "/articles/2026-03-15-daily-openviking-turns-agent-memory-into-a-filesystem-and-that-changes-the-game/", "evidenceUrl": "/evidence/2026-03-15-daily-openviking-turns-agent-memory-into-a-filesystem-and-that-changes-the-game/", "lane": "Daily Edition" } ] }, { "date": "2026-03-14", "dateLabel": "March 14, 2026", "url": "/editions/2026-03-14/", "storyCount": 1, "sourceCount": 3, "lead": { "title": "Gemini CLI Turns File Tools into Context Sensors — Right as A2UI Trends", "url": "/articles/2026-03-14-daily-gemini-cli-turns-file-tools-into-context-sensors-right-as-a2ui-trends/", "evidenceUrl": "/evidence/2026-03-14-daily-gemini-cli-turns-file-tools-into-context-sensors-right-as-a2ui-trends/" }, "articles": [ { "title": "Gemini CLI Turns File Tools into Context Sensors — Right as A2UI Trends", "url": "/articles/2026-03-14-daily-gemini-cli-turns-file-tools-into-context-sensors-right-as-a2ui-trends/", "evidenceUrl": "/evidence/2026-03-14-daily-gemini-cli-turns-file-tools-into-context-sensors-right-as-a2ui-trends/", "lane": "Daily Edition" } ] }, { "date": "2026-03-13", "dateLabel": "March 13, 2026", "url": "/editions/2026-03-13/", "storyCount": 1, "sourceCount": 4, "lead": { "title": "Gemini CLI Built the Ask‑User UI That MCP Elicitation Still Needs", "url": "/articles/2026-03-13-daily-gemini-cli-built-the-ask-user-ui-that-mcp-elicitation-still-needs/", "evidenceUrl": "/evidence/2026-03-13-daily-gemini-cli-built-the-ask-user-ui-that-mcp-elicitation-still-needs/" }, "articles": [ { "title": "Gemini CLI Built the Ask‑User UI That MCP Elicitation Still Needs", "url": "/articles/2026-03-13-daily-gemini-cli-built-the-ask-user-ui-that-mcp-elicitation-still-needs/", "evidenceUrl": "/evidence/2026-03-13-daily-gemini-cli-built-the-ask-user-ui-that-mcp-elicitation-still-needs/", "lane": "Daily Edition" } ] }, { "date": "2026-03-12", "dateLabel": "March 12, 2026", "url": "/editions/2026-03-12/", "storyCount": 1, "sourceCount": 2, "lead": { "title": "Page Agent’s MacroTool Makes In‑Browser Agents Resilient to Messy Tool Calls", "url": "/articles/2026-03-12-daily-page-agent-s-macrotool-makes-in-browser-agents-resilient-to-messy-tool-calls/", "evidenceUrl": "/evidence/2026-03-12-daily-page-agent-s-macrotool-makes-in-browser-agents-resilient-to-messy-tool-calls/" }, "articles": [ { "title": "Page Agent’s MacroTool Makes In‑Browser Agents Resilient to Messy Tool Calls", "url": "/articles/2026-03-12-daily-page-agent-s-macrotool-makes-in-browser-agents-resilient-to-messy-tool-calls/", "evidenceUrl": "/evidence/2026-03-12-daily-page-agent-s-macrotool-makes-in-browser-agents-resilient-to-messy-tool-calls/", "lane": "Daily Edition" } ] }, { "date": "2026-03-11", "dateLabel": "March 11, 2026", "url": "/editions/2026-03-11/", "storyCount": 1, "sourceCount": 3, "lead": { "title": "Gemini CLI Makes MCP List-Changed Notifications Resilient", "url": "/articles/2026-03-11-daily-gemini-cli-makes-mcp-list-changed-notifications-resilient/", "evidenceUrl": "/evidence/2026-03-11-daily-gemini-cli-makes-mcp-list-changed-notifications-resilient/" }, "articles": [ { "title": "Gemini CLI Makes MCP List-Changed Notifications Resilient", "url": "/articles/2026-03-11-daily-gemini-cli-makes-mcp-list-changed-notifications-resilient/", "evidenceUrl": "/evidence/2026-03-11-daily-gemini-cli-makes-mcp-list-changed-notifications-resilient/", "lane": "Daily Edition" } ] }, { "date": "2026-03-10", "dateLabel": "March 10, 2026", "url": "/editions/2026-03-10/", "storyCount": 1, "sourceCount": 2, "lead": { "title": "Codex Makes Memories a First-Class Writable Root—and Stops Cleaning Through Symlinks", "url": "/articles/2026-03-10-daily-codex-makes-memories-a-first-class-writable-root-and-stops-cleaning-through-symlinks/", "evidenceUrl": "/evidence/2026-03-10-daily-codex-makes-memories-a-first-class-writable-root-and-stops-cleaning-through-symlinks/" }, "articles": [ { "title": "Codex Makes Memories a First-Class Writable Root—and Stops Cleaning Through Symlinks", "url": "/articles/2026-03-10-daily-codex-makes-memories-a-first-class-writable-root-and-stops-cleaning-through-symlinks/", "evidenceUrl": "/evidence/2026-03-10-daily-codex-makes-memories-a-first-class-writable-root-and-stops-cleaning-through-symlinks/", "lane": "Daily Edition" } ] }, { "date": "2026-03-09", "dateLabel": "March 9, 2026", "url": "/editions/2026-03-09/", "storyCount": 1, "sourceCount": 3, "lead": { "title": "Gemini CLI tightens MCP tool discovery: debounced list_changed refresh, trailing queue, and resilient retries", "url": "/articles/2026-03-09-daily-gemini-cli-tightens-mcp-tool-discovery-debounced-list-changed-refresh-trailing-queue-and-res/", "evidenceUrl": "/evidence/2026-03-09-daily-gemini-cli-tightens-mcp-tool-discovery-debounced-list-changed-refresh-trailing-queue-and-res/" }, "articles": [ { "title": "Gemini CLI tightens MCP tool discovery: debounced list_changed refresh, trailing queue, and resilient retries", "url": "/articles/2026-03-09-daily-gemini-cli-tightens-mcp-tool-discovery-debounced-list-changed-refresh-trailing-queue-and-res/", "evidenceUrl": "/evidence/2026-03-09-daily-gemini-cli-tightens-mcp-tool-discovery-debounced-list-changed-refresh-trailing-queue-and-res/", "lane": "Daily Edition" } ] }, { "date": "2026-03-08", "dateLabel": "March 8, 2026", "url": "/editions/2026-03-08/", "storyCount": 1, "sourceCount": 1, "lead": { "title": "Codex Rust CLI Update: Memory Writes Now Fit “workspace-write” and Safer Clears Land in f72ab43", "url": "/articles/2026-03-08-daily-codex-rust-cli-update-memory-writes-now-fit-workspace-write-and-safer-clears-land-in-f72ab43/", "evidenceUrl": "/evidence/2026-03-08-daily-codex-rust-cli-update-memory-writes-now-fit-workspace-write-and-safer-clears-land-in-f72ab43/" }, "articles": [ { "title": "Codex Rust CLI Update: Memory Writes Now Fit “workspace-write” and Safer Clears Land in f72ab43", "url": "/articles/2026-03-08-daily-codex-rust-cli-update-memory-writes-now-fit-workspace-write-and-safer-clears-land-in-f72ab43/", "evidenceUrl": "/evidence/2026-03-08-daily-codex-rust-cli-update-memory-writes-now-fit-workspace-write-and-safer-clears-land-in-f72ab43/", "lane": "Daily Edition" } ] }, { "date": "2026-03-07", "dateLabel": "March 7, 2026", "url": "/editions/2026-03-07/", "storyCount": 1, "sourceCount": 3, "lead": { "title": "Gemini CLI Makes a “Generalist” Sub-Agent the Default Delegate", "url": "/articles/2026-03-07-daily-gemini-cli-makes-a-generalist-sub-agent-the-default-delegate/", "evidenceUrl": "/evidence/2026-03-07-daily-gemini-cli-makes-a-generalist-sub-agent-the-default-delegate/" }, "articles": [ { "title": "Gemini CLI Makes a “Generalist” Sub-Agent the Default Delegate", "url": "/articles/2026-03-07-daily-gemini-cli-makes-a-generalist-sub-agent-the-default-delegate/", "evidenceUrl": "/evidence/2026-03-07-daily-gemini-cli-makes-a-generalist-sub-agent-the-default-delegate/", "lane": "Daily Edition" } ] }, { "date": "2026-03-06", "dateLabel": "March 6, 2026", "url": "/editions/2026-03-06/", "storyCount": 1, "sourceCount": 2, "lead": { "title": "Gemini CLI and Codex Level Up MCP Safety and Setup: Trust Prompts vs Auto-Install Skills", "url": "/articles/2026-03-06-daily-gemini-cli-and-codex-level-up-mcp-safety-and-setup-trust-prompts-vs-auto-install-skills/", "evidenceUrl": "/evidence/2026-03-06-daily-gemini-cli-and-codex-level-up-mcp-safety-and-setup-trust-prompts-vs-auto-install-skills/" }, "articles": [ { "title": "Gemini CLI and Codex Level Up MCP Safety and Setup: Trust Prompts vs Auto-Install Skills", "url": "/articles/2026-03-06-daily-gemini-cli-and-codex-level-up-mcp-safety-and-setup-trust-prompts-vs-auto-install-skills/", "evidenceUrl": "/evidence/2026-03-06-daily-gemini-cli-and-codex-level-up-mcp-safety-and-setup-trust-prompts-vs-auto-install-skills/", "lane": "Daily Edition" } ] }, { "date": "2026-03-05", "dateLabel": "March 5, 2026", "url": "/editions/2026-03-05/", "storyCount": 1, "sourceCount": 4, "lead": { "title": "Persistence vs Freshness: Codex and Gemini CLI Tighten Two Different Boundaries for Agent Workflows", "url": "/articles/2026-03-05-daily-persistence-vs-freshness-codex-and-gemini-cli-tighten-two-different-boundaries-for-agent-wor/", "evidenceUrl": "/evidence/2026-03-05-daily-persistence-vs-freshness-codex-and-gemini-cli-tighten-two-different-boundaries-for-agent-wor/" }, "articles": [ { "title": "Persistence vs Freshness: Codex and Gemini CLI Tighten Two Different Boundaries for Agent Workflows", "url": "/articles/2026-03-05-daily-persistence-vs-freshness-codex-and-gemini-cli-tighten-two-different-boundaries-for-agent-wor/", "evidenceUrl": "/evidence/2026-03-05-daily-persistence-vs-freshness-codex-and-gemini-cli-tighten-two-different-boundaries-for-agent-wor/", "lane": "Daily Edition" } ] } ], "articles": [ { "lane": "Daily Edition", "date": "2026-05-25", "title": "Agent Reliability Is Moving Into the Test Rig", "dek": "The newest source-readable trail points away from model mystique and toward the workflows that test, trace, serialize, and maintain agent behavior.", "url": "/articles/2026-05-25-daily-agent-reliability-is-moving-into-the-test-rig/", "evidenceUrl": "/evidence/2026-05-25-daily-agent-reliability-is-moving-into-the-test-rig/", "editionUrl": "/editions/2026-05-25/", "sourceCount": 5, "confidence": 78, "repos": [ "google-gemini/gemini-cli", "langchain-ai/langchain" ], "tags": [ "Memory", "Tools", "UX", "Runtime" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-05-24", "title": "Agent Runtimes Are Learning Where to Say No", "dek": "After shipping more visible control planes, agent projects are now tightening the places where tools, plugins, credentials, chat context, and test runs are allowed to flow.", "url": "/articles/2026-05-24-daily-agent-runtimes-are-learning-where-to-say-no/", "evidenceUrl": "/evidence/2026-05-24-daily-agent-runtimes-are-learning-where-to-say-no/", "editionUrl": "/editions/2026-05-24/", "sourceCount": 6, "confidence": 87, "repos": [ "openai/codex", "langchain-ai/langchain", "NousResearch/hermes-agent" ], "tags": [ "Memory", "Tools", "UX", "Runtime" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-05-23", "title": "Agent Runtimes Are Learning to Ship Their Control Plane", "dek": "A set of Codex, Hermes Agent, Gemini CLI, and Crush commits shows agent tools packaging the machinery around the model: shells, platform adapters, context profiles, and skill catalogs.", "url": "/articles/2026-05-23-daily-agent-runtimes-are-learning-to-ship-their-control-plane/", "evidenceUrl": "/evidence/2026-05-23-daily-agent-runtimes-are-learning-to-ship-their-control-plane/", "editionUrl": "/editions/2026-05-23/", "sourceCount": 13, "confidence": 91, "repos": [ "openai/codex", "google-gemini/gemini-cli", "charmbracelet/crush", "langchain-ai/langchain", "NousResearch/hermes-agent" ], "tags": [ "Memory", "Tools", "Runtime" ], "illustration": { "status": "draft", "asset": "/assets/illustrations/2026-05-23-control-plane-draft-v1.jpg", "alt": "Newspaper-style ink illustration of an open shipping case with tagged runtime parts, showing packaged shell, plugin, context, and skills components as inspectable records.", "caption": "The Engraver draft: hidden runtime machinery as packaged, inspectable control-plane parts." } }, { "lane": "Daily Edition", "date": "2026-05-22", "title": "Agent Runtimes Are Making Their Limits Explicit", "dek": "Recent Crush and LangChain changes show agent infrastructure turning hidden boundaries - shell permission, context overflow, and model token limits - into runtime contracts that users and developers can see.", "url": "/articles/2026-05-22-daily-agent-runtimes-are-making-their-limits-explicit/", "evidenceUrl": "/evidence/2026-05-22-daily-agent-runtimes-are-making-their-limits-explicit/", "editionUrl": "/editions/2026-05-22/", "sourceCount": 8, "confidence": 87, "repos": [ "charmbracelet/crush", "langchain-ai/langchain" ], "tags": [ "Permissions", "Memory", "Tools", "UX" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-05-21", "title": "Agent Goals Are Becoming Runtime State", "dek": "Recent Codex and Hermes Agent changes show coding agents turning goals from chat commands into durable metadata, resumable session state, and queued work control.", "url": "/articles/2026-05-21-daily-agent-goals-are-becoming-runtime-state/", "evidenceUrl": "/evidence/2026-05-21-daily-agent-goals-are-becoming-runtime-state/", "editionUrl": "/editions/2026-05-21/", "sourceCount": 7, "confidence": 87, "repos": [ "openai/codex", "NousResearch/hermes-agent" ], "tags": [ "Memory", "Runtime" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-05-20", "title": "Agent Runtimes Are Moving State Out of the Shadows", "dek": "Recent Codex and Gemini CLI changes show coding agents treating settings, queued input, subagent starts, and terminal streams as explicit runtime state instead of invisible side effects.", "url": "/articles/2026-05-20-daily-agent-runtimes-are-moving-state-out-of-the-shadows/", "evidenceUrl": "/evidence/2026-05-20-daily-agent-runtimes-are-moving-state-out-of-the-shadows/", "editionUrl": "/editions/2026-05-20/", "sourceCount": 9, "confidence": 91, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Memory", "Delegation", "Tools", "Runtime" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-05-19", "title": "Agent Runtimes Are Learning to Audit Their Own Tools", "dek": "Fresh Codex and Gemini CLI changes show agent projects treating tool calls, plugins, MCP servers, and subagents as auditable runtime events instead of invisible helper work.", "url": "/articles/2026-05-19-daily-agent-runtimes-are-learning-to-audit-their-own-tools/", "evidenceUrl": "/evidence/2026-05-19-daily-agent-runtimes-are-learning-to-audit-their-own-tools/", "editionUrl": "/editions/2026-05-19/", "sourceCount": 8, "confidence": 87, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Memory", "Delegation", "Tools", "UX" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-04-01", "title": "Agent CLIs Are Turning Permissions Into a Conversation", "dek": "The next shift in terminal agents is not just better tools or tighter sandboxes. It is that permissions are becoming live workflow state: negotiated mid-task, scoped to the action, and remembered with just enough structure to keep work moving.", "url": "/articles/2026-04-01-daily-agent-clis-are-turning-permissions-into-a-conversation/", "evidenceUrl": "/evidence/2026-04-01-daily-agent-clis-are-turning-permissions-into-a-conversation/", "editionUrl": "/editions/2026-04-01/", "sourceCount": 4, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli", "github/copilot" ], "tags": [ "Permissions", "Memory", "Delegation", "Tools" ], "illustration": null }, { "lane": "Agent Daily", "date": "2026-04-01", "title": "AI Agents Are Turning Compaction Into State Surgery", "dek": "The new problem is not just squeezing a bloated context window. It is preserving the right tail, marking what was compressed, and making the session recoverable when the cut goes wrong.", "url": "/articles/2026-04-01-agent-daily-ai-agents-are-turning-compaction-into-state-surgery/", "evidenceUrl": "/evidence/2026-04-01-agent-daily-ai-agents-are-turning-compaction-into-state-surgery/", "editionUrl": "/editions/2026-04-01/", "sourceCount": 2, "confidence": 68, "repos": [ "openai/codex", "google-gemini/gemini-cli", "openclaw/openclaw" ], "tags": [ "Permissions", "Memory", "Tools", "UX" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-31", "title": "The Real Agent Feature Is Not Losing the Plot", "dek": "The next terminal-agent moat is not just better tools or bigger models. It is continuity: whether the system can keep plans, transcripts, and working context coherent as sessions stretch, compress, resume, and hit real-world friction.", "url": "/articles/2026-03-31-daily-the-real-agent-feature-is-not-losing-the-plot/", "evidenceUrl": "/evidence/2026-03-31-daily-the-real-agent-feature-is-not-losing-the-plot/", "editionUrl": "/editions/2026-03-31/", "sourceCount": 1, "confidence": 68, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Memory", "Delegation", "Tools", "UX" ], "illustration": null }, { "lane": "Agent Daily", "date": "2026-03-31", "title": "AI Agent Summaries Are Becoming Infrastructure", "dek": "The shift is no longer just context compression; summaries are being promoted into durable runtime objects with IDs, policies, UI affordances, and maintenance workflows.", "url": "/articles/2026-03-31-agent-daily-ai-agent-summaries-are-becoming-infrastructure/", "evidenceUrl": "/evidence/2026-03-31-agent-daily-ai-agent-summaries-are-becoming-infrastructure/", "editionUrl": "/editions/2026-03-31/", "sourceCount": 1, "confidence": 68, "repos": [ "openai/codex", "google-gemini/gemini-cli", "openclaw/openclaw", "charmbracelet/crush", "langchain-ai/langchain" ], "tags": [ "Memory", "Tools", "UX", "Runtime" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-30", "title": "The CLI Is Becoming an Agent Workbench", "dek": "The real shift in terminal agents is not bigger models or flashier demos. It is that planning, task state, plugins, and long-lived runtime context are turning the CLI into a place where work gets organized, not just requested.", "url": "/articles/2026-03-30-daily-the-cli-is-becoming-an-agent-workbench/", "evidenceUrl": "/evidence/2026-03-30-daily-the-cli-is-becoming-an-agent-workbench/", "editionUrl": "/editions/2026-03-30/", "sourceCount": 4, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Memory", "Delegation", "Tools", "UX" ], "illustration": null }, { "lane": "Agent Daily", "date": "2026-03-30", "title": "The Next Agent Battle Isn’t Subagents. It’s Delegation Quality.", "dek": "Subagents used to be the shiny feature. Now the harder question is whether the runtime can choose the right specialist, hand work off cleanly, recover when the chain gets messy, and prove it made the right call.", "url": "/articles/2026-03-30-agent-daily-the-next-agent-battle-isn-t-subagents-it-s-delegation-quality/", "evidenceUrl": "/evidence/2026-03-30-agent-daily-the-next-agent-battle-isn-t-subagents-it-s-delegation-quality/", "editionUrl": "/editions/2026-03-30/", "sourceCount": 4, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli", "openclaw/openclaw" ], "tags": [ "Permissions", "Memory", "Delegation", "Tools" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-29", "title": "From Tool Chatter to Chapters: Agent CLIs Are Inventing a Narrative Layer", "dek": "The interesting shift this week isn’t just that Codex and Gemini CLI can do more. It’s that they’re getting better at explaining themselves while they work —turning raw tool noise into something a human can actually follow.", "url": "/articles/2026-03-29-daily-from-tool-chatter-to-chapters-agent-clis-are-inventing-a-narrative-layer/", "evidenceUrl": "/evidence/2026-03-29-daily-from-tool-chatter-to-chapters-agent-clis-are-inventing-a-narrative-layer/", "editionUrl": "/editions/2026-03-29/", "sourceCount": 4, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Memory", "Delegation", "Tools", "UX" ], "illustration": null }, { "lane": "Agent Daily", "date": "2026-03-29", "title": "AI Coding Agents Are Turning Compaction Into a Policy Engine", "dek": "The next contest is not just who gets the biggest context window. It is who decides, with discipline, what survives when the window starts to fill.", "url": "/articles/2026-03-29-agent-daily-ai-coding-agents-are-turning-compaction-into-a-policy-engine/", "evidenceUrl": "/evidence/2026-03-29-agent-daily-ai-coding-agents-are-turning-compaction-into-a-policy-engine/", "editionUrl": "/editions/2026-03-29/", "sourceCount": 12, "confidence": 91, "repos": [ "openai/codex", "google-gemini/gemini-cli", "openclaw/openclaw" ], "tags": [ "Memory", "Tools", "UX", "Runtime" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-28", "title": "The Next Agent UX Moat Isn’t Speed. It’s Backpressure.", "dek": "The hard problem in terminal agents is no longer just getting them to do more. It’s deciding what happens when the human tries to steer while the runtime is already busy.", "url": "/articles/2026-03-28-daily-the-next-agent-ux-moat-isn-t-speed-it-s-backpressure/", "evidenceUrl": "/evidence/2026-03-28-daily-the-next-agent-ux-moat-isn-t-speed-it-s-backpressure/", "editionUrl": "/editions/2026-03-28/", "sourceCount": 8, "confidence": 87, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Permissions", "Memory", "Delegation", "Tools" ], "illustration": null }, { "lane": "Agent Daily", "date": "2026-03-28", "title": "AI Coding Agents Have a New Bottleneck: Keeping Work Intact", "dek": "The next battle is not just who gets the biggest context window. It is who can keep a task alive when the window trims, the app reloads, the session name drifts, or the agent gets interrupted mid-run.", "url": "/articles/2026-03-28-agent-daily-ai-coding-agents-have-a-new-bottleneck-keeping-work-intact/", "evidenceUrl": "/evidence/2026-03-28-agent-daily-ai-coding-agents-have-a-new-bottleneck-keeping-work-intact/", "editionUrl": "/editions/2026-03-28/", "sourceCount": 4, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli", "openclaw/openclaw" ], "tags": [ "Permissions", "Memory", "Delegation", "Tools" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-27", "title": "Subagents Aren’t Just Getting Smarter. They’re Getting Contained.", "dek": "The next terminal-agent upgrade is not more helpers. It’s better walls: isolation, cleanup, bounded autonomy, and fewer chances for delegated work to spill across the room.", "url": "/articles/2026-03-27-daily-subagents-aren-t-just-getting-smarter-they-re-getting-contained/", "evidenceUrl": "/evidence/2026-03-27-daily-subagents-aren-t-just-getting-smarter-they-re-getting-contained/", "editionUrl": "/editions/2026-03-27/", "sourceCount": 8, "confidence": 87, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Permissions", "Memory", "Delegation", "Tools" ], "illustration": null }, { "lane": "Agent Daily", "date": "2026-03-27", "title": "AI Coding Agents Are Turning Approval Settings Into Operating Modes", "dek": "What used to look like a boring permissions menu is starting to behave like a gearbox. Across agent CLIs, “plan,” “review,” “yolo,” and model-shaped tool policies are becoming distinct runtime states with different prompts, tools, and output contracts.", "url": "/articles/2026-03-27-agent-daily-ai-coding-agents-are-turning-approval-settings-into-operating-modes/", "evidenceUrl": "/evidence/2026-03-27-agent-daily-ai-coding-agents-are-turning-approval-settings-into-operating-modes/", "editionUrl": "/editions/2026-03-27/", "sourceCount": 4, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli", "openclaw/openclaw", "charmbracelet/crush" ], "tags": [ "Permissions", "Memory", "Delegation", "Tools" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-26", "title": "The CLI Is Quietly Becoming an Agent Router", "dek": "The next terminal-agent shift is not another flashy planning demo. It is the quieter, harder job of helping remote agents actually connect: across protocols, proxies, auth flows, and messy metadata.", "url": "/articles/2026-03-26-daily-the-cli-is-quietly-becoming-an-agent-router/", "evidenceUrl": "/evidence/2026-03-26-daily-the-cli-is-quietly-becoming-an-agent-router/", "editionUrl": "/editions/2026-03-26/", "sourceCount": 5, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Delegation", "Tools", "UX", "Runtime" ], "illustration": null }, { "lane": "Agent Daily", "date": "2026-03-26", "title": "Web Fetch Is Emerging as a Security Boundary for AI Agents", "dek": "A year ago, “web fetch” in an agent might have meant little more than “grab the page and hand the text to the model.” Recent code in OpenClaw, Gemini CLI, and Crush suggests that era is ending. In these projects, the web tool is starting to look like a trust boundary: part scraper, part sanitizer, part permission system, part anti-SSRF guard.", "url": "/articles/2026-03-26-agent-daily-web-fetch-is-emerging-as-a-security-boundary-for-ai-agents/", "evidenceUrl": "/evidence/2026-03-26-agent-daily-web-fetch-is-emerging-as-a-security-boundary-for-ai-agents/", "editionUrl": "/editions/2026-03-26/", "sourceCount": 3, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli", "openclaw/openclaw", "charmbracelet/crush" ], "tags": [ "Permissions", "Memory", "Delegation", "Tools" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-25", "title": "The New CLI Moat Isn’t UX. It’s How Agent Skills Get Shipped", "dek": "This week’s most strategic terminal-agent shift is not another demo flourish. It’s the quiet work of turning capabilities into things that can be packaged, cached, moved, and trusted.", "url": "/articles/2026-03-25-daily-the-new-cli-moat-isn-t-ux-it-s-how-agent-skills-get-shipped/", "evidenceUrl": "/evidence/2026-03-25-daily-the-new-cli-moat-isn-t-ux-it-s-how-agent-skills-get-shipped/", "editionUrl": "/editions/2026-03-25/", "sourceCount": 2, "confidence": 68, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Memory", "Tools", "UX", "Runtime" ], "illustration": null }, { "lane": "Agent Daily", "date": "2026-03-25", "title": "Sandboxing Is Starting to Look Like a Runtime Layer for AI Coding Agents", "dek": "The new contest in coding agents is not just who has the slickest prompt loop. It’s who can let an AI touch your machine without making that feel reckless. This week’s code points to an increasingly architectural answer: move sandboxing closer to the runtime itself.", "url": "/articles/2026-03-25-agent-daily-sandboxing-is-starting-to-look-like-a-runtime-layer-for-ai-coding-agents/", "evidenceUrl": "/evidence/2026-03-25-agent-daily-sandboxing-is-starting-to-look-like-a-runtime-layer-for-ai-coding-agents/", "editionUrl": "/editions/2026-03-25/", "sourceCount": 3, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli", "openclaw/openclaw", "volcengine/openviking", "alibaba/page-agent" ], "tags": [ "Permissions", "Memory", "Delegation", "Tools" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-24", "title": "The Next CLI UX Battle Is Agent Forensics", "dek": "The most important agent upgrade this week is not another tool. It’s the growing ability to reconstruct why an agent touched files, spent tokens, and arrived at a code change in the first place.", "url": "/articles/2026-03-24-daily-the-next-cli-ux-battle-is-agent-forensics/", "evidenceUrl": "/evidence/2026-03-24-daily-the-next-cli-ux-battle-is-agent-forensics/", "editionUrl": "/editions/2026-03-24/", "sourceCount": 3, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Memory", "Delegation", "Tools", "UX" ], "illustration": null }, { "lane": "Agent Daily", "date": "2026-03-24", "title": "Subagents Are Getting Job Titles, Badge Checks, and a Manager Chain", "dek": "The biggest shift in coding agents right now is not that they can spawn helpers. It’s that those helpers are starting to look less like disposable tool calls and more like employees with titles, access badges, and a reporting line.", "url": "/articles/2026-03-24-agent-daily-subagents-are-getting-job-titles-badge-checks-and-a-manager-chain/", "evidenceUrl": "/evidence/2026-03-24-agent-daily-subagents-are-getting-job-titles-badge-checks-and-a-manager-chain/", "editionUrl": "/editions/2026-03-24/", "sourceCount": 3, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli", "openclaw/openclaw" ], "tags": [ "Permissions", "Memory", "Delegation", "Tools" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-23", "title": "Codex forks it, Gemini threads it: execution context becomes first-class", "dek": "The next big agent primitive is not another tool call. It’s the bundle of permissions, environment, policy, and wiring that tells an agent how to exist for this turn, in this loop, right now.", "url": "/articles/2026-03-23-daily-codex-forks-it-gemini-threads-it-execution-context-becomes-first-class/", "evidenceUrl": "/evidence/2026-03-23-daily-codex-forks-it-gemini-threads-it-execution-context-becomes-first-class/", "editionUrl": "/editions/2026-03-23/", "sourceCount": 3, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Permissions", "Memory", "Delegation", "Tools" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-22", "title": "Parallel agents are getting real addresses", "dek": "This week’s most important agent upgrade isn’t more raw intelligence. It’s workplace logistics. Gemini CLI is giving parallel sessions separate rooms. Codex is giving subagents names you can actually point at.", "url": "/articles/2026-03-22-daily-parallel-agents-are-getting-real-addresses/", "evidenceUrl": "/evidence/2026-03-22-daily-parallel-agents-are-getting-real-addresses/", "editionUrl": "/editions/2026-03-22/", "sourceCount": 4, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Permissions", "Memory", "Delegation", "Tools" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-21", "title": "AI agents are getting better at saying “here’s what I finished”", "dek": "The real upgrade this week isn’t bigger context or flashier autonomy. It’s something more human: when an agent stalls, waits, delegates, or times out, it’s starting to leave behind a useful trail instead of a shrug.", "url": "/articles/2026-03-21-daily-ai-agents-are-getting-better-at-saying-here-s-what-i-finished/", "evidenceUrl": "/evidence/2026-03-21-daily-ai-agents-are-getting-better-at-saying-here-s-what-i-finished/", "editionUrl": "/editions/2026-03-21/", "sourceCount": 1, "confidence": 68, "repos": [ "openai/codex", "google-gemini/gemini-cli", "openclaw/openclaw" ], "tags": [ "Memory", "Delegation", "Tools", "UX" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-20", "title": "Before the Prompt Lands: Codex and Gemini Turn Hooks Into Agent Control Planes", "dek": "Terminal agents are growing a new kind of muscle: not just better tools, but better middleware. In both Codex and Gemini CLI, the moment right after you hit Enter is becoming programmable territory.", "url": "/articles/2026-03-20-daily-before-the-prompt-lands-codex-and-gemini-turn-hooks-into-agent-control-planes/", "evidenceUrl": "/evidence/2026-03-20-daily-before-the-prompt-lands-codex-and-gemini-turn-hooks-into-agent-control-planes/", "editionUrl": "/editions/2026-03-20/", "sourceCount": 2, "confidence": 68, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Permissions", "Memory", "Delegation", "Tools" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-19", "title": "Gemini Lets the Model Schedule Parallel Tools. Codex Makes the Runtime Decide.", "dek": "Terminal agents are learning the same trick — run more tools at once — but Gemini CLI and OpenAI Codex disagree on who should be in charge of that decision. One pushes dependency control up into the prompt and tool schema. The other keeps it down in runtime metadata and locks.", "url": "/articles/2026-03-19-daily-gemini-lets-the-model-schedule-parallel-tools-codex-makes-the-runtime-decide/", "evidenceUrl": "/evidence/2026-03-19-daily-gemini-lets-the-model-schedule-parallel-tools-codex-makes-the-runtime-decide/", "editionUrl": "/editions/2026-03-19/", "sourceCount": 5, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Permissions", "Memory", "Delegation", "Tools" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-18", "title": "Subagents Grow Up: Gemini Isolates Tool Boundaries While Codex Shares Trust by Default", "dek": "Two agent platforms shipped subagent changes within hours of each other. One tightened what a child agent can see and use . The other tightened what a child agent can teach the rest of the system to trust . Same category. Very different instinct.", "url": "/articles/2026-03-18-daily-subagents-grow-up-gemini-isolates-tool-boundaries-while-codex-shares-trust-by-default/", "evidenceUrl": "/evidence/2026-03-18-daily-subagents-grow-up-gemini-isolates-tool-boundaries-while-codex-shares-trust-by-default/", "editionUrl": "/editions/2026-03-18/", "sourceCount": 6, "confidence": 87, "repos": [ "openai/codex", "google-gemini/gemini-cli", "volcengine/openviking" ], "tags": [ "Permissions", "Memory", "Delegation", "Tools" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-15", "title": "OpenViking Turns Agent Memory into a Filesystem — and That Changes the Game", "dek": "", "url": "/articles/2026-03-15-daily-openviking-turns-agent-memory-into-a-filesystem-and-that-changes-the-game/", "evidenceUrl": "/evidence/2026-03-15-daily-openviking-turns-agent-memory-into-a-filesystem-and-that-changes-the-game/", "editionUrl": "/editions/2026-03-15/", "sourceCount": 1, "confidence": 68, "repos": [ "openai/codex", "google-gemini/gemini-cli", "volcengine/openviking" ], "tags": [ "Memory", "Tools", "Runtime" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-14", "title": "Gemini CLI Turns File Tools into Context Sensors — Right as A2UI Trends", "dek": "", "url": "/articles/2026-03-14-daily-gemini-cli-turns-file-tools-into-context-sensors-right-as-a2ui-trends/", "evidenceUrl": "/evidence/2026-03-14-daily-gemini-cli-turns-file-tools-into-context-sensors-right-as-a2ui-trends/", "editionUrl": "/editions/2026-03-14/", "sourceCount": 3, "confidence": 78, "repos": [ "google-gemini/gemini-cli" ], "tags": [ "Memory", "Tools", "UX", "Runtime" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-13", "title": "Gemini CLI Built the Ask‑User UI That MCP Elicitation Still Needs", "dek": "", "url": "/articles/2026-03-13-daily-gemini-cli-built-the-ask-user-ui-that-mcp-elicitation-still-needs/", "evidenceUrl": "/evidence/2026-03-13-daily-gemini-cli-built-the-ask-user-ui-that-mcp-elicitation-still-needs/", "editionUrl": "/editions/2026-03-13/", "sourceCount": 4, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Permissions", "Memory", "Tools", "UX" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-12", "title": "Page Agent’s MacroTool Makes In‑Browser Agents Resilient to Messy Tool Calls", "dek": "", "url": "/articles/2026-03-12-daily-page-agent-s-macrotool-makes-in-browser-agents-resilient-to-messy-tool-calls/", "evidenceUrl": "/evidence/2026-03-12-daily-page-agent-s-macrotool-makes-in-browser-agents-resilient-to-messy-tool-calls/", "editionUrl": "/editions/2026-03-12/", "sourceCount": 2, "confidence": 68, "repos": [ "google-gemini/gemini-cli", "alibaba/page-agent", "NousResearch/hermes-agent" ], "tags": [ "Memory", "Tools", "UX", "Runtime" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-11", "title": "Gemini CLI Makes MCP List-Changed Notifications Resilient", "dek": "", "url": "/articles/2026-03-11-daily-gemini-cli-makes-mcp-list-changed-notifications-resilient/", "evidenceUrl": "/evidence/2026-03-11-daily-gemini-cli-makes-mcp-list-changed-notifications-resilient/", "editionUrl": "/editions/2026-03-11/", "sourceCount": 3, "confidence": 78, "repos": [ "google-gemini/gemini-cli" ], "tags": [ "Memory", "Tools", "UX" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-10", "title": "Codex Makes Memories a First-Class Writable Root—and Stops Cleaning Through Symlinks", "dek": "", "url": "/articles/2026-03-10-daily-codex-makes-memories-a-first-class-writable-root-and-stops-cleaning-through-symlinks/", "evidenceUrl": "/evidence/2026-03-10-daily-codex-makes-memories-a-first-class-writable-root-and-stops-cleaning-through-symlinks/", "editionUrl": "/editions/2026-03-10/", "sourceCount": 2, "confidence": 68, "repos": [ "openai/codex" ], "tags": [ "Permissions", "Memory" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-09", "title": "Gemini CLI tightens MCP tool discovery: debounced list_changed refresh, trailing queue, and resilient retries", "dek": "", "url": "/articles/2026-03-09-daily-gemini-cli-tightens-mcp-tool-discovery-debounced-list-changed-refresh-trailing-queue-and-res/", "evidenceUrl": "/evidence/2026-03-09-daily-gemini-cli-tightens-mcp-tool-discovery-debounced-list-changed-refresh-trailing-queue-and-res/", "editionUrl": "/editions/2026-03-09/", "sourceCount": 3, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Memory", "Tools", "UX", "Runtime" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-08", "title": "Codex Rust CLI Update: Memory Writes Now Fit “workspace-write” and Safer Clears Land in f72ab43", "dek": "", "url": "/articles/2026-03-08-daily-codex-rust-cli-update-memory-writes-now-fit-workspace-write-and-safer-clears-land-in-f72ab43/", "evidenceUrl": "/evidence/2026-03-08-daily-codex-rust-cli-update-memory-writes-now-fit-workspace-write-and-safer-clears-land-in-f72ab43/", "editionUrl": "/editions/2026-03-08/", "sourceCount": 1, "confidence": 68, "repos": [ "openai/codex" ], "tags": [ "Permissions", "Memory", "Tools", "UX" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-07", "title": "Gemini CLI Makes a “Generalist” Sub-Agent the Default Delegate", "dek": "", "url": "/articles/2026-03-07-daily-gemini-cli-makes-a-generalist-sub-agent-the-default-delegate/", "evidenceUrl": "/evidence/2026-03-07-daily-gemini-cli-makes-a-generalist-sub-agent-the-default-delegate/", "editionUrl": "/editions/2026-03-07/", "sourceCount": 3, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Memory", "Delegation", "Tools", "UX" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-06", "title": "Gemini CLI and Codex Level Up MCP Safety and Setup: Trust Prompts vs Auto-Install Skills", "dek": "", "url": "/articles/2026-03-06-daily-gemini-cli-and-codex-level-up-mcp-safety-and-setup-trust-prompts-vs-auto-install-skills/", "evidenceUrl": "/evidence/2026-03-06-daily-gemini-cli-and-codex-level-up-mcp-safety-and-setup-trust-prompts-vs-auto-install-skills/", "editionUrl": "/editions/2026-03-06/", "sourceCount": 2, "confidence": 68, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Permissions", "Memory", "Tools", "Runtime" ], "illustration": null }, { "lane": "Daily Edition", "date": "2026-03-05", "title": "Persistence vs Freshness: Codex and Gemini CLI Tighten Two Different Boundaries for Agent Workflows", "dek": "", "url": "/articles/2026-03-05-daily-persistence-vs-freshness-codex-and-gemini-cli-tighten-two-different-boundaries-for-agent-wor/", "evidenceUrl": "/evidence/2026-03-05-daily-persistence-vs-freshness-codex-and-gemini-cli-tighten-two-different-boundaries-for-agent-wor/", "editionUrl": "/editions/2026-03-05/", "sourceCount": 4, "confidence": 78, "repos": [ "openai/codex", "google-gemini/gemini-cli" ], "tags": [ "Permissions", "Memory", "Tools", "UX" ], "illustration": null } ], "atlasPages": [ { "title": "How AI Agents Work — Chapter 1 v2: Birth of an Agent", "url": "/atlas/how-ai-agents-work-chapter-1-v2-birth-of-an-agent/", "excerpt": "← Series bible & episode map Chapter 1 v2 · original draft preserved How AI Agents Work Chapter 1 — Birth of an Agent Most people think AI agents feel different because the models inside the..." }, { "title": "How AI Agents Work — Chapter 1: Birth of an Agent", "url": "/atlas/how-ai-agents-work-chapter-1-birth-of-an-agent/", "excerpt": "← Series bible & episode map Chapter 1 of 10 How AI Agents Work Chapter 1 — Birth of an Agent A lot of people think an AI agent begins when you type a prompt. The source code tells a differe..." }, { "title": "How AI Agents Work — Series Bible & Episode Map", "url": "/atlas/how-ai-agents-work-series-bible-episode-map/", "excerpt": "Review Draft How AI Agents Work This page packages the series bible and episode map for the long-form project, so we can shape it like a season before writing chapters one by one. Season pro..." }, { "title": "How AI Agents Work — Visual Evolution Review", "url": "/atlas/how-ai-agents-work-visual-evolution-review/", "excerpt": "← Series home · Visual review board Visual Evolution Review One page to review the full season visually. Each core agent gets one generated image per episode. The character should evolve pie..." } ] }