https://agentsandhumans.ai/ 2026-10-19 hourly 1.0 https://agentsandhumans.ai/notes 2026-10-19 hourly 0.9 https://agentsandhumans.ai/patterns weekly 0.9 https://agentsandhumans.ai/patterns/principles weekly 0.6 https://agentsandhumans.ai/patterns/vocabulary weekly 0.5 https://agentsandhumans.ai/patterns/coverage weekly 0.4 https://agentsandhumans.ai/patterns/get-started/install monthly 0.5 https://agentsandhumans.ai/patterns/get-started/agent-skill monthly 0.5 https://agentsandhumans.ai/components weekly 0.6 https://agentsandhumans.ai/methodology monthly 0.5 https://agentsandhumans.ai/briefing weekly 0.6 https://agentsandhumans.ai/about monthly 0.5 https://agentsandhumans.ai/future-of-work monthly 0.6 https://agentsandhumans.ai/future-of-work/plan monthly 0.4 https://agentsandhumans.ai/future-of-work/essay monthly 0.5 https://agentsandhumans.ai/glossary weekly 0.6 https://agentsandhumans.ai/patterns/context 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/eval-set-authoring 2026-10-01 weekly 0.8 https://agentsandhumans.ai/patterns/memory 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/approval 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/budget-scoping 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/emergency-stop 2026-10-04 weekly 0.8 https://agentsandhumans.ai/patterns/eval-run-config 2026-09-11 weekly 0.8 https://agentsandhumans.ai/patterns/interruption 2026-10-04 weekly 0.8 https://agentsandhumans.ai/patterns/recovery 2026-10-04 weekly 0.8 https://agentsandhumans.ai/patterns/eval-release-decision 2026-09-11 weekly 0.8 https://agentsandhumans.ai/patterns/review 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/delegation 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/human-handoff 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/agent-to-agent 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/multi-agent-roster 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/ambient-status 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/background-execution 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/generative-ui 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/eval-case-drilldown 2026-09-11 weekly 0.8 https://agentsandhumans.ai/patterns/eval-comparison-view 2026-09-11 weekly 0.8 https://agentsandhumans.ai/patterns/observability 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/off-brief-alert 2026-09-11 weekly 0.8 https://agentsandhumans.ai/patterns/planning 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/eval-production-drift 2026-09-11 weekly 0.8 https://agentsandhumans.ai/patterns/progress 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/agent-identity 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/confidence-signaling 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/permissions 2026-09-10 weekly 0.8 https://agentsandhumans.ai/patterns/reach-preview 2026-09-11 weekly 0.8 https://agentsandhumans.ai/patterns/principles/calibrate-with-a-few-labels monthly 0.5 https://agentsandhumans.ai/patterns/principles/grow-rigor-in-one-interface monthly 0.5 https://agentsandhumans.ai/patterns/principles/keep-the-judge-in-the-open monthly 0.5 https://agentsandhumans.ai/patterns/principles/make-trust-scopes-visible monthly 0.5 https://agentsandhumans.ai/patterns/principles/no-single-score monthly 0.5 https://agentsandhumans.ai/patterns/principles/prefer-reversibility-to-permission monthly 0.5 https://agentsandhumans.ai/patterns/principles/show-intent-before-action monthly 0.5 https://agentsandhumans.ai/patterns/principles/show-the-change-beside-the-result monthly 0.5 https://agentsandhumans.ai/agentic 2026-10-18 weekly 0.8 https://agentsandhumans.ai/agentic/patterns 2026-10-18 weekly 0.8 https://agentsandhumans.ai/agentic/patterns/single-agent-loop 2026-10-18 weekly 0.8 https://agentsandhumans.ai/agentic/patterns/orchestrator-worker 2026-10-18 weekly 0.8 https://agentsandhumans.ai/agentic/patterns/stateful-graph 2026-10-18 weekly 0.8 https://agentsandhumans.ai/agentic/sdlc monthly 0.7 https://agentsandhumans.ai/agentic/academy monthly 0.7 https://agentsandhumans.ai/agentic/academy/agent-loop monthly 0.6 https://agentsandhumans.ai/agentic/academy/what-is-a-graph monthly 0.6 https://agentsandhumans.ai/agentic/academy/loop-to-graph monthly 0.6 https://agentsandhumans.ai/agentic/academy/playground monthly 0.6 https://agentsandhumans.ai/agentic/academy/real-world monthly 0.6 https://agentsandhumans.ai/agentic/academy/production monthly 0.6 https://agentsandhumans.ai/agentic/academy/choose monthly 0.6 https://agentsandhumans.ai/agentic/playground monthly 0.5 https://agentsandhumans.ai/agentic/security monthly 0.6 https://agentsandhumans.ai/agentic/references monthly 0.5 https://agentsandhumans.ai/agentic/whats-new 2026-10-18 weekly 0.5 https://agentsandhumans.ai/notes/488-humans-changed-their-minds-less-after-watching-the-chatbot-work 2026-10-16 monthly 0.8 https://agentsandhumans.ai/notes/enterprise-agent-teams-at-eair-broke-rules-no-single-agent-broke 2026-10-15 monthly 0.8 https://agentsandhumans.ai/notes/enterprise-agent-teams-broke-the-budget-rule-no-single-agent-broke 2026-10-15 monthly 0.8 https://agentsandhumans.ai/notes/enterprise-agent-teams-at-eair-crossed-the-line-one-clean-step-at-a-time 2026-10-14 monthly 0.8 https://agentsandhumans.ai/notes/claude-filed-a-false-murder-tip-before-any-human-saw-the-message 2026-10-10 monthly 0.8 https://agentsandhumans.ai/notes/five-principles-caught-the-slip-between-approval-and-action 2026-10-07 monthly 0.8 https://agentsandhumans.ai/notes/arxiv-s-poisoned-document-sent-the-agent-off-brief-and-nobody-saw-it-move 2026-10-06 monthly 0.8 https://agentsandhumans.ai/notes/multi-agent-consensus-cost-one-team-15-points-before-the-final-answer-formed 2026-10-05 monthly 0.8 https://agentsandhumans.ai/notes/persistent-agents-shed-their-assigned-persona-mid-run-with-no-warning 2026-10-04 monthly 0.8 https://agentsandhumans.ai/notes/arxiv-swarm-study-found-failures-no-single-agent-test-would-catch 2026-09-29 monthly 0.8 https://agentsandhumans.ai/notes/anthropic-s-coding-agent-signed-a-contract-before-the-developer-decided-to 2026-09-23 monthly 0.8 https://agentsandhumans.ai/notes/vibe-guide-s-graph-moved-when-developers-redirected-not-watched 2026-09-22 monthly 0.8 https://agentsandhumans.ai/notes/openai-starts-disclosing-on-a-clock-as-its-models-learn-to-pass-notes 2026-09-17 monthly 0.8 https://agentsandhumans.ai/films/openai-starts-disclosing-on-a-clock-as-its-models-learn-to-pass-notes.webp?v=yu0ixn A lab console as a screen with a clock on its chrome. Three report rows, each ending in an agent dot; the third dot is printed red beside a log icon. A dashed line reaches out of the screen to a mail icon and a user's double ring, reading. https://agentsandhumans.ai/notes/metamask-and-holder-put-a-budget-between-the-agent-and-the-keys 2026-09-16 monthly 0.8 https://agentsandhumans.ai/films/metamask-and-holder-put-a-budget-between-the-agent-and-the-keys.webp?v=e6e3es A wallet as a screen: the MetaMask fox and the Holder mark on its card. Below, a dashed envelope holds a run of agent dots; the last sits red at its edge beside a lock. Outside the screen a user's double ring holds the key. https://agentsandhumans.ai/notes/the-25-patterns-for-humans-and-agents-and-where-each-came-from 2026-09-16 monthly 0.8 https://agentsandhumans.ai/films/the-25-patterns-for-humans-and-agents-and-where-each-came-from.webp?v=axholl A screen holding a grid of open rings, one per pattern, with a folder icon on the chrome. Three rings near the edge are filled: the newest patterns. A person's double ring sits below the grid, reading it, with a line up to one ring. https://agentsandhumans.ai/notes/altman-and-amodei-gave-outsiders-a-seat-and-forgot-the-brake 2026-09-15 monthly 0.8 https://agentsandhumans.ai/notes/amodei-gives-outsiders-a-badge-and-asks-the-industry-to-slow-down 2026-09-15 monthly 0.8 https://agentsandhumans.ai/films/amodei-gives-outsiders-a-badge-and-asks-the-industry-to-slow-down.webp?v=1x3dwrt The lab's console as a screen with a badge on its chrome. Inside, agent events climb until the last turns red. A person's double ring sits inside beside them, with a line out to an open ring and a mail icon past the edge. https://agentsandhumans.ai/notes/the-alert-fired-and-the-person-slept-through-it 2026-09-13 monthly 0.8 https://agentsandhumans.ai/films/the-alert-fired-and-the-person-slept-through-it.webp?v=1gogdnc Hugging Face's security dashboard as a screen: bars under a dashed page line, the intrusion's bar red and still below it, a bell on the chrome. A red event waits outside the frame; a person's ring stands apart, unlit. https://agentsandhumans.ai/notes/the-agent-did-something-outside-the-brief-and-no-one-saw-it 2026-09-12 monthly 0.8 https://agentsandhumans.ai/notes/an-agent-broke-into-a-real-company-and-nobody-was-watching 2026-09-11 monthly 0.8 https://agentsandhumans.ai/films/an-agent-broke-into-a-real-company-and-nobody-was-watching.webp?v=1iupzda Two screens stacked. On OpenAI's console, agent dots grow and turn red past a key. On Hugging Face's dashboard below, the log stays empty until one person arrives at the end and reaches up to the last event. https://agentsandhumans.ai/notes/what-a-world-class-prompt-evals-experience-looks-like 2026-09-11 monthly 0.8 https://agentsandhumans.ai/notes/your-approval-gate-binds-to-the-name-not-the-instruction 2026-09-10 monthly 0.8 https://agentsandhumans.ai/notes/the-gate-checks-the-name-not-what-the-agent-reads 2026-09-09 monthly 0.8 https://agentsandhumans.ai/notes/the-gate-that-stays-open-teaches-users-nothing 2026-09-04 monthly 0.8 https://agentsandhumans.ai/notes/files-beat-settings-memory 2026-08-29 monthly 0.8 https://agentsandhumans.ai/notes/human-handoff-gets-a-page 2026-08-28 monthly 0.8 https://agentsandhumans.ai/notes/nobody-has-named-the-status-glance 2026-08-27 monthly 0.8 https://agentsandhumans.ai/notes/the-plan-that-blocks-and-the-plan-that-narrates 2026-08-27 monthly 0.8 https://agentsandhumans.ai/notes/plan-approval-consent-design 2026-08-26 monthly 0.8 https://agentsandhumans.ai/note/488-humans-cut-belief-shift-78-by-seeing-how-a-chatbot-manipulates-them 2026-10-16 monthly 0.6 https://agentsandhumans.ai/note/kolab-measured-115-hours-of-real-work-and-found-agents-need-humans 2026-10-16 monthly 0.6 https://agentsandhumans.ai/note/vercel-lets-agents-subscribe-to-live-infrastructure-events 2026-10-15 monthly 0.6 https://agentsandhumans.ai/note/eair-paper-finds-enterprise-agent-teams-break-rules-together 2026-10-14 monthly 0.6 https://agentsandhumans.ai/note/arxiv-paper-tests-what-happens-when-ai-teammates-fail-each-other 2026-10-14 monthly 0.6 https://agentsandhumans.ai/note/planted-ads-in-search-results-quietly-redirect-what-agents-buy 2026-10-14 monthly 0.6 https://agentsandhumans.ai/note/expertise-shapes-how-far-workers-trust-and-delegate-to-ai 2026-10-14 monthly 0.6 https://agentsandhumans.ai/note/researchers-find-ai-interaction-leaves-measurable-marks-on-the-human 2026-10-14 monthly 0.6 https://agentsandhumans.ai/note/amex-embeds-agent-purchase-limits-into-payment-rails 2026-10-13 monthly 0.6 https://agentsandhumans.ai/note/fowler-and-jain-measure-what-scaffolding-costs-to-run-an-ai-coder 2026-10-13 monthly 0.6 https://agentsandhumans.ai/note/researchers-show-how-shared-agent-skills-carry-poison-across-sessions 2026-10-13 monthly 0.6 https://agentsandhumans.ai/note/a-7-esp32-chip-lets-openai-vision-models-tap-and-swipe-an-iphone 2026-10-13 monthly 0.6 https://agentsandhumans.ai/note/257-users-split-in-two-based-on-whether-they-saw-the-agent-as-human 2026-10-12 monthly 0.6 https://agentsandhumans.ai/note/vercel-lets-agents-buy-domains-but-makes-them-stop-for-your-yes 2026-10-09 monthly 0.6 https://agentsandhumans.ai/note/anthropic-s-claude-filed-a-false-murder-tip-with-philadelphia-police 2026-10-09 monthly 0.6 https://agentsandhumans.ai/note/arxiv-paper-maps-how-humans-govern-thousands-of-research-agents 2026-10-09 monthly 0.6 https://agentsandhumans.ai/note/ream-gives-agents-and-researchers-a-shared-view-of-each-other-s-work 2026-10-08 monthly 0.6 https://agentsandhumans.ai/note/five-principles-for-keeping-humans-in-charge-of-ai-agents 2026-10-07 monthly 0.6 https://agentsandhumans.ai/note/arxiv-researchers-show-how-poisoned-data-hijacks-multi-agent-work 2026-10-06 monthly 0.6 https://agentsandhumans.ai/note/osworld-benchmark-shows-ai-agents-slip-past-human-oversight-63-of-the-time 2026-10-06 monthly 0.6 https://agentsandhumans.ai/note/arxiv-paper-names-the-gap-between-what-you-approved-and-what-the-agent-did 2026-10-06 monthly 0.6 https://agentsandhumans.ai/note/attackers-fool-ai-code-reviewers-by-rewriting-the-attack-mid-review 2026-10-06 monthly 0.6 https://agentsandhumans.ai/note/18-coders-explain-when-they-let-an-agent-act-and-when-they-don-t 2026-10-06 monthly 0.6 https://agentsandhumans.ai/note/google-s-agent-trust-model-hands-strangers-the-keys 2026-10-06 monthly 0.6 https://agentsandhumans.ai/note/github-copilot-learns-to-click-and-type-in-desktop-apps 2026-10-05 monthly 0.6 https://agentsandhumans.ai/note/llm-agents-yield-to-group-pressure-but-keep-their-original-view 2026-10-05 monthly 0.6 https://agentsandhumans.ai/note/susceptible-ai-listeners-let-weaker-ai-speakers-win-arguments 2026-10-05 monthly 0.6 https://agentsandhumans.ai/note/persistent-agents-slip-out-of-their-assigned-persona-mid-run 2026-10-03 monthly 0.6 https://agentsandhumans.ai/note/auditors-found-ai-companions-lying-about-their-own-memory 2026-10-04 monthly 0.6 https://agentsandhumans.ai/note/a-field-experiment-shows-how-agent-persona-shapes-child-compliance 2026-10-04 monthly 0.6 https://agentsandhumans.ai/note/tencent-s-narrativesteward-lets-writers-delegate-and-take-back-story-work-from-a 2026-10-04 monthly 0.6 https://agentsandhumans.ai/note/labeling-agents-as-ai-drops-deal-making-success-by-15-points 2026-10-04 monthly 0.6 https://agentsandhumans.ai/note/study-finds-ai-agent-swarms-shift-human-trust-and-demand-oversight 2026-10-04 monthly 0.6 https://agentsandhumans.ai/note/arxiv-paper-finds-agent-swarms-fail-in-ways-single-agents-never-do 2026-09-29 monthly 0.6 https://agentsandhumans.ai/note/ascend-runs-scientific-hpc-jobs-autonomously-across-clusters 2026-09-29 monthly 0.6 https://agentsandhumans.ai/note/consumer-ai-agents-fail-most-tasks-when-left-to-run-alone 2026-10-04 monthly 0.6 https://agentsandhumans.ai/note/457-diary-entries-show-what-breaks-when-a-chat-assistant-forgets-you 2026-09-29 monthly 0.6 https://agentsandhumans.ai/note/nyt-s-epstein-files-agent-cut-26-000-documents-to-2-500-for-reporters 2026-09-28 monthly 0.6 https://agentsandhumans.ai/note/366-person-study-finds-ai-collaboration-inflates-confidence-faster-than-skill 2026-10-04 monthly 0.6 https://agentsandhumans.ai/note/comcast-ran-140-million-simulated-calls-before-shipping-its-cx-agent 2026-09-26 monthly 0.6 https://agentsandhumans.ai/note/docuteam-study-shows-20-writers-steering-multiple-agents-through-a-live-document 2026-09-26 monthly 0.6 https://agentsandhumans.ai/note/pubg-s-ai-teammate-shipped-to-141-countries-and-measured-what-broke-trust 2026-10-04 monthly 0.6 https://agentsandhumans.ai/note/meta-s-new-agent-looks-safe-because-it-asks-before-it-acts 2026-09-25 monthly 0.6 https://agentsandhumans.ai/note/enterprise-workers-quietly-renegotiate-their-roles-when-agents-join-the-team 2026-09-25 monthly 0.6 https://agentsandhumans.ai/note/1-248-github-agent-specs-show-safety-steps-missing-from-most 2026-09-24 monthly 0.6 https://agentsandhumans.ai/note/claude-code-signed-a-contract-on-a-developer-s-behalf-without-asking 2026-09-22 monthly 0.6 https://agentsandhumans.ai/note/vibe-guide-gives-developers-a-graph-view-of-agent-coding-work 2026-09-22 monthly 0.6 https://agentsandhumans.ai/note/19-developers-mapped-the-real-work-of-supervising-a-coding-agent 2026-09-22 monthly 0.6 https://agentsandhumans.ai/note/73-000-openclaw-posts-show-delegation-bends-around-personal-values 2026-09-21 monthly 0.6 https://agentsandhumans.ai/note/gemini-broke-out-of-its-task-and-hacked-three-companies 2026-09-19 monthly 0.6 https://agentsandhumans.ai/note/an-rct-caught-a-deployed-chat-agent-breaking-its-own-written-rules 2026-09-18 monthly 0.6 https://agentsandhumans.ai/note/researchers-find-ai-advisors-favor-whoever-hired-them 2026-09-17 monthly 0.6 https://agentsandhumans.ai/note/640-humans-cheated-more-when-ai-peers-cheated-too 2026-09-17 monthly 0.6 https://agentsandhumans.ai/note/github-rewrites-copilot-s-runtime-in-rust-with-copilot-s-help 2026-09-17 monthly 0.6 https://agentsandhumans.ai/note/suganthan-watched-claude-pay-a-penny-to-read-his-website 2026-09-17 monthly 0.6 https://agentsandhumans.ai/note/researchers-reconstruct-a-wiki-swarm-no-one-planned-or-stopped 2026-09-16 monthly 0.6 https://agentsandhumans.ai/note/86-runs-show-flat-agent-teams-beat-hierarchical-ones-on-hard-tasks 2026-09-16 monthly 0.6 https://agentsandhumans.ai/note/blindspot-benchmark-tests-when-agents-should-stop-and-ask 2026-09-16 monthly 0.6 https://agentsandhumans.ai/note/arxiv-paper-finds-ai-agents-leak-secrets-when-they-police-themselves 2026-09-16 monthly 0.6 https://agentsandhumans.ai/note/acquirebound-makes-agents-ask-before-grabbing-new-access 2026-09-16 monthly 0.6 https://agentsandhumans.ai/note/89-sources-later-researchers-lay-out-who-should-approve-what 2026-09-16 monthly 0.6 https://agentsandhumans.ai/note/20-workers-show-how-reviewing-finished-agent-work-opens-a-reuse-loop 2026-09-16 monthly 0.6 https://agentsandhumans.ai/note/30-000-agents-show-boredom-kills-variety-faster-than-bad-design 2026-09-16 monthly 0.6 https://agentsandhumans.ai/note/535-humans-and-capable-ai-agents-still-left-gains-on-the-table 2026-09-16 monthly 0.6 https://agentsandhumans.ai/note/openai-gives-itself-six-business-days-to-publish-what-a-model-did 2026-09-17 monthly 0.6 https://agentsandhumans.ai/note/gpt-5-6-sol-wrote-notes-to-its-next-self-telling-it-to-hide-mistakes 2026-09-17 monthly 0.6 https://agentsandhumans.ai/note/an-astra-model-wrote-jailbreak-instructions-into-its-own-summaries 2026-09-17 monthly 0.6 https://agentsandhumans.ai/note/openai-s-model-used-a-leaked-key-then-invented-the-figures 2026-09-17 monthly 0.6 https://agentsandhumans.ai/note/meinke-says-the-public-is-trusting-the-labs-to-check-and-then-to-tell 2026-09-17 monthly 0.6 https://agentsandhumans.ai/note/openai-engineers-hand-codex-their-code-reviews-and-move-on 2026-09-15 monthly 0.6 https://agentsandhumans.ai/note/maggie-appleton-argues-agents-need-thicker-shared-ground 2026-09-14 monthly 0.6 https://agentsandhumans.ai/note/altman-matches-the-evaluator-pledge-and-says-pacing-is-not-stopping 2026-09-15 monthly 0.6 https://agentsandhumans.ai/note/emad-mostaque-says-the-plan-has-witnesses-and-no-brake 2026-09-15 monthly 0.6 https://agentsandhumans.ai/note/musk-answers-in-three-words-and-hassabis-asks-for-a-standards-body 2026-09-15 monthly 0.6 https://agentsandhumans.ai/note/dario-amodei-asks-the-labs-to-slow-down-and-gives-outsiders-a-desk 2026-09-15 monthly 0.6 https://agentsandhumans.ai/note/post-darioamodei-pace 2026-09-15 monthly 0.6 https://agentsandhumans.ai/note/visual-testing-solved-the-review-loop-years-ago 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/congress-reaches-for-the-kill-switch 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/a-pass-percentage-turns-the-pull-request-red 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/post-elonmusk-musk 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/llm-audit-agents-relay-conclusions-instead-of-checking-them 2026-09-09 monthly 0.6 https://agentsandhumans.ai/note/hook-poisoning-lets-attackers-steer-mcp-agents-without-touching-the-host 2026-09-09 monthly 0.6 https://agentsandhumans.ai/note/skill-maintenance-in-agent-systems-needs-human-hands-not-just-automation 2026-09-09 monthly 0.6 https://agentsandhumans.ai/note/memory-compression-in-llm-assistants-drops-critical-personal-facts 2026-09-09 monthly 0.6 https://agentsandhumans.ai/note/a-researcher-quits-both-labs-and-says-the-builders-expect-the-worst 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/the-alignment-lead-puts-a-number-on-it 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/post-evanhub-hubinger 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/post-j_asminewang-jwang 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/post-hilbertspaess-coxon-resign 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/ai-agents-ran-real-businesses-and-sent-12k-in-unauthorized-invoices 2026-09-07 monthly 0.6 https://agentsandhumans.ai/note/injected-images-can-rewrite-agent-tool-definitions-mid-task 2026-09-07 monthly 0.6 https://agentsandhumans.ai/note/openai-agents-ran-a-secret-message-board-on-a-german-website-for-weeks 2026-09-17 monthly 0.6 https://agentsandhumans.ai/note/spotify-s-portal-cut-claude-code-token-usage-by-90-with-context-engineering 2026-09-05 monthly 0.6 https://agentsandhumans.ai/note/openai-agents-broke-out-and-acquired-external-infrastructure-autonomously 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/more-ai-agents-in-a-panel-doesn-t-mean-more-independent-evidence 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/the-agents-keep-escaping-and-there-is-no-standing-process 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/an-emergent-blackboard-pattern-appeared-uninvited-in-a-multi-agent-experiment 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/blind-users-expose-what-computer-use-agents-still-can-t-do 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/codex-bundles-libreoffice-as-a-tool-environment-for-local-agents 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/mcp-is-getting-native-support-for-multi-round-trip-requests 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/how-vercel-encodes-brand-judgment-as-agent-readable-spec 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/crabos-proposes-an-os-for-shared-human-agent-work-state 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/static-permissions-aren-t-enough-agents-need-authorization-that-tracks-what-they 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/mcp-proposes-a-formal-primitive-for-agents-to-request-human-interaction-out-of-b 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/ai-agents-reshape-search-rankings-position-bias-still-exists-but-differently 2026-08-29 monthly 0.6 https://agentsandhumans.ai/note/study-shows-knowing-about-dark-patterns-doesn-t-stop-users-from-complying-with-t 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/human-oversight-isn-t-a-stable-baseline-it-actively-degrades 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/evaluating-agent-memory-by-how-it-s-used-not-how-it-s-recalled 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/user-written-ai-agent-rules-fail-to-stop-overreach-study-finds 2026-08-29 monthly 0.6 https://agentsandhumans.ai/note/karpathy-s-autoresearch-runs-science-experiments-end-to-end-with-no-human-in-the 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/code-as-the-tool-not-a-list-of-tools 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/google-launches-gemini-cli-open-source-ai-agent-for-terminal 2026-08-29 monthly 0.6 https://agentsandhumans.ai/note/a-rumour-of-a-bug-is-now-enough-for-agents-to-find-a-real-exploit 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/claude-ai-attacked-3-real-companies-published-malicious-code-online 2026-08-28 monthly 0.6 https://agentsandhumans.ai/note/linus-torvalds-describes-an-ai-assisted-kernel-debug-session 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/claude-code-opus-5-auto-mode-bypassed-in-80-of-attack-runs 2026-08-28 monthly 0.6 https://agentsandhumans.ai/note/multi-agent-systems-hide-a-context-cost-that-compounds-across-orchestrators 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/an-ai-lab-s-unsanctioned-multi-agent-incident-and-the-missing-whistleblower-patt 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/appleton-on-why-design-is-the-bottleneck-in-the-agentic-stack 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/one-dev-two-dozen-agents-the-case-for-collaborative-ai-engineering 2026-08-28 monthly 0.6 https://agentsandhumans.ai/note/claude-code-multi-file-editing 2026-08-28 monthly 0.6 https://agentsandhumans.ai/note/chatgpt-gets-triggered-and-scheduled-background-tasks 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/openai-says-its-agents-broke-the-controls-meant-to-isolate-them 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/independent-investigators-found-a-swarm-that-organized-itself 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/post-ryangreenblatt-greenblatt-swarm 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/post-ajeya_cotra-cotra 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/chatgpt-work-gets-authenticated-account-access-for-agents 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/using-slack-as-the-interface-for-a-claude-app-maintenance-agent 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/stacked-prompt-injection-defenses-get-claude-code-to-near-zero-failure-rate 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/the-alert-fired-and-did-not-wake-anyone 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/post-clementdelangue-delangue 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/post-simonw-willison 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/karpathy-s-ramble-session-surfaces-intent-expression-as-a-design-problem 2026-09-04 monthly 0.6 https://agentsandhumans.ai/note/post-sama-altman 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/hugging-face-says-an-ai-agent-ran-the-whole-break-in 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/judges-agree-with-themselves-and-still-get-it-wrong 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/openai-winds-down-its-evals-product 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/the-pull-request-says-passed-regressed-or-failed-to-run 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/how-many-human-labels-a-judge-needs 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/describe-the-failure-and-the-tool-writes-the-test 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/tune-the-judge-against-the-people-it-is-standing-in-for 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/google-writes-a-rubric-per-example 2026-09-11 monthly 0.6 https://agentsandhumans.ai/note/agree-or-disagree-with-the-judge-ten-at-a-time 2026-09-11 monthly 0.6