◉CLAUDE LABJP
●PRICING — Anthropic has cancelled the September 1 increase for Claude Sonnet 5. The $2 input and $10 output per million tokens is now simply the standard price, so any forecast built on $3 and $15 needs redoing●VERSION — v2.1.241, released August 23, is a fixes-and-reliability release. MCP v2 no longer reopens subscriptions/listen endlessly against servers that close long-held streams on a fixed timeout●LINUX — Idle sessions on Linux with sandboxing enabled no longer pin a CPU core at 100 percent●SKILLS — Bundled skill aliases such as /checkup and /review no longer report Unknown command in -p mode or with plugins and MCP loaded when a same-named user or project skill shadows them●ARGS — Skill and command argument substitution no longer re-expands argument values as template markers●WATERMARK — Claude products released from August 2 onward embed machine-readable marking in generated output. It answers EU AI Act transparency rules, but it is applied worldwide rather than only in Europe●PRICING — Anthropic has cancelled the September 1 increase for Claude Sonnet 5. The $2 input and $10 output per million tokens is now simply the standard price, so any forecast built on $3 and $15 needs redoing●VERSION — v2.1.241, released August 23, is a fixes-and-reliability release. MCP v2 no longer reopens subscriptions/listen endlessly against servers that close long-held streams on a fixed timeout●LINUX — Idle sessions on Linux with sandboxing enabled no longer pin a CPU core at 100 percent●SKILLS — Bundled skill aliases such as /checkup and /review no longer report Unknown command in -p mode or with plugins and MCP loaded when a same-named user or project skill shadows them●ARGS — Skill and command argument substitution no longer re-expands argument values as template markers●WATERMARK — Claude products released from August 2 onward embed machine-readable marking in generated output. It answers EU AI Act transparency rules, but it is applied worldwide rather than only in Europe
TAG

Token Optimization

3 articles
← Back to all tags
Related:
Claude API3Agent SDK1MCP1Context Management1tool use1vision1tool_result1context editing1memory tool1agents1
⬡ API & SDK/2026-06-30Advanced

When a Tool Result Is Too Big and Melts Your Context Window: Designing Cursor-Based Pagination

When a list tool returns hundreds of rows at once, an agent's context can collapse in a single call. Here is a cursor-based pagination design that keeps tool output small and protects your token budget, with working code.

⬡ API & SDK/2026-06-29Advanced

Let Claude Actually See the Images Your Tools Return — Use Image Blocks in tool_result and Cut Tokens by Roughly 10x

Stuffing a base64 string into a tool_result makes the same image cost roughly 10–20x more tokens. Here is how to return it as an image content block instead, with SDK code, a token-cost estimate, and the gotchas I hit in production.

⬡ API & SDK/2026-06-16Advanced

Taming Token Bloat in Long-Running Agents with Context Editing and the Memory Tool

For long-running agents whose input tokens balloon as tool results pile up, here is how to pair context editing with the memory tool and measure the savings with count_tokens, including a working backend implementation.