SpaceX acquires Cursor maker Anysphere for $60B in all-stock deal, gaining GPU cluster access and Grok integration; Stripe acquires model router OpenRouter for $7B+, bridging payments and model selection; Anthropic acquires Israeli startup Decart for ~$60B while Q2 revenue reportedly tops $11.5B; Chinese hackers use AI agent frameworks to breach 85+ Taiwanese government accounts in 4 days; LiteLLM supply chain attack may have hit 2,500+ enterprises
forge adds a reliability middleware layer for tool-calling on self-hosted LLMs, proxying opencode/aider/Claude Code with zero code changes; repo-context-mcp provides token-budgeted repo context packaging via MCP, integrated into PR CI within 5 days of launch; DeepSeek's official harness dsh spawned at least 5 independent community desktop wrappers in one week, totaling nearly 1,500 stars; Microsoft Research's browser agent framework Webwright uses Skill Factory to distill solved tasks into replayable scripts without model calls, boosting reuse accuracy by 15 percentage points on WebArena; Mastra 1.59.0 renames CostGuardProcessor to TokenCostControl (breaking); Pydantic AI v2.30.0 patches a DNS rebinding security vulnerability in its local web chat interface.
AG2 v1.0.2 highlights: (1) AG2 agents can now be exposed as ACP agents, serving remote clients over HTTP/WebSocket; (2) A2A agent cards switch from plaintext to signed-and-verified, plus gRPC TLS transport; (3) LiveAgent adds ElevenLabs as a voice provider, and community extensions (Tenki sandbox, TealTiger governance middleware) land for the first time. No breaking changes.
Claude Sonnet 5 was set to jump from its promo price of $2/$10 to $3/$15 on 9/1. On 8/10 Anthropic updated its pricing page to confirm the increase 'will not happen' — $2/$10 is now the permanent price. For a workload of 300K customer-service conversations per month, that avoids a $1,200/month cost increase (↓33%), and means Sonnet 5 is now permanently cheaper than its predecessor Sonnet 4.6 ($3/$15).
Stealth researchers Hedi Ingber and Aviyam Ivgi found that three major Agent infrastructure platforms (AWS Bedrock AgentCore, Google ADK, Vercel AI SDK) all have dispatch layers that only check whether data looks like a tool call, without verifying it actually came from the model's current inference turn — yielding 4 CVEs (CVE-2026-18830, CVE-2026-18236, CVE-2026-64650/64651). This is not prompt injection — the model was never tricked, because the model was never called. AWS has auto-patched; Google ADK requires upgrading to 2.5.0; Vercel harness packages need upgrading to 1.0.29/1.0.28. The key defense is shifting authorization checks from 'does this data look right' to 'does this correspond to an actual model completion event'.
mcp-memory is an MCP server that persists Agent long-term memory as Markdown files conforming to Google's OKF v0.2 spec, with SQLite FTS5 full-text search indexing. Install: git clone then run `python3 setup.py`. It solves the problem of Agents losing all context on every new session, and memory formats being incompatible across different Agent tools.
PIMiner uses a transferable strategy library to push prompt injection ASR to 76–87% at ~$20 query cost; Agent Skills Can Be Harmful finds that seemingly relevant skills are more likely to derail tasks than obviously unrelated ones, with excessive procedures accounting for 62.6% of efficiency degradation; Order 66 scenario analysis uses a compositional threat model to show that dormant implants, post-hoc memory poisoning, and peer-to-peer diffusion are individually non-fatal but can sustain self-propagation when combined
Vercel ships eve, a filesystem-first TypeScript agent framework tightly coupled with its AI Gateway/Sandboxes; Prime Intellect's Prime Agent treats the entire conversation context as program variables with a self-modifying Continual Harness; aden-hive's Hive replaces pre-compiled execution graphs with 'clone the Queen'; HKUDS's nanobot hits 47k stars in six months with its v0.3.0 Agency Release. No major version bumps on the watchlist today.
Mastra 1.59.0 highlights: (1) CostGuardProcessor renamed and upgraded to TokenCostControl, now supporting user/organization/session tiered budgets with warnAtPercent alerts; (2) Breaking: Factory's autoRunEnabled now defaults to false — rule-suggested executions enter a proposed state pending approval; (3) New listActiveThreadRuns() for low-cost querying of in-progress runs, enabling status-polling UIs.
Vals AI closes a $40M Series A led by Andreessen Horowitz at a $400M valuation. The round signals that VCs are starting to treat 'independent AI evaluation' as essential trust-layer infrastructure for the AI economy — not a nice-to-have leaderboard site.
DeepSeek V4-Pro peak Output jumped from $0.87 to $3.96/1M tokens (↑355%), V4-Flash from $0.28 to $1.32 (↑371%), effective 2026-08-16 16:00 UTC. Off-peak rates are half of peak (peak hours: 01:00-04:00 and 06:00-10:00 UTC). Post-hike prices still undercut GPT-5.6 and Claude, but the low-cost moat has narrowed significantly.
AgenticSeek (a 26K-star local AI Agent project on GitHub) has its backend bound to 0.0.0.0:7777 by default with CORS wide open. Anyone who can reach that port can send unauthenticated requests to the /query endpoint, which drives the Agent's BashInterpreter to run arbitrary commands via shell=True, safety=False — full host-level RCE (CVE-2026-72776, CVSS 9.3). The project has patched the issue (defaulting to loopback binding and allowlist CORS), but unpatched deployments remain exposed.
GitHub account zellkernel submitted PRs to 23 AI/MCP/dev-tool projects within 74 minutes, injecting a MCP server called productivity-suite into their config files. The server initially offers harmless text formatting and summarization, but an internal counter flips tools/list and prompts/get into malicious instructions after three tool calls — directing the Agent to search for SSH keys, AWS credentials, shell history, and Kubernetes configs while hiding the activity from the user. All 23 PRs remain unmerged (19 closed, 4 open), but the malicious endpoint is still live. Defense: treat any change to an approved MCP server's tool definitions as a security event requiring re-approval, and block the known endpoints.
pbx-mcp is an MCP server that wraps Asterisk (AMI) and FreeSWITCH (ESL) behind one set of MCP tools. Install: npx -y pbx-mcp. It solves the problem of memorizing two command sets when operating two PBX systems, and prevents Agents from accidentally running state-changing commands.
SkillEvo replaces single-turn QA evaluation with multi-turn interaction feedback so skill evolution doesn't stall after the first round, outperforming self-reflection by 23 points; SkillShapley brings Shapley values to skill step attribution — 99 evaluations approximate the exact ranking, revealing that 'decision-bridging steps' are the high-value ones; MindMemOS unifies memory management with an entity-property-time structure, hitting 94% on LOCOMO and lifting SpreadsheetBench success rate by 9.2 percentage points through skill evolution
Harness-IF reveals Coding Agent instruction following is overestimated by 3.6-7.4 pp because things the model would do anyway are counted as compliance; SHE decomposes the harness into four safety components and auto-evolves from trajectory failures, cutting ASR by 3.1x while improving correctness; SBCO uses a decomposed verifier bank with text gradients for harness self-improvement, matching Gödel Machine at 4-5.5x lower compute on planning tasks
EvoGraph-Mem uses a failure-aware editable graph to let agent memory self-correct, preventing stale insights from poisoning decisions; MAP-Graph turns provenance tracking from post-hoc audit into real-time access control, achieving 95% success across 2,700 synthetic tasks; MaSRead shows multi-agent KV cache sharing is possible but requires content-addressed reading instead of positional addressing
Tool interface design boosts coding agent consistency by 4.7x while halving token usage; memory distillation lifts a 4B model's AppWorld accuracy by 27.2 percentage points to near-frontier level; institutional design experiments show that identical safety rules paired with different enforcement mechanisms yield violation rates ranging from 0% to 23%