Users report Cursor forcing fast mode on composer 2.5 despite disabling it
A Reddit user reports that Cursor's composer 2.5 is running in fast mode even after they disabled it in settings and subagents, and added rules to prevent fast mode. The user notes that fast mode consumes 6x more usage, which is a concern for those on limited budgets.
Entities
Related
Cursor users report worsening UI changes and recurring bugs
A Reddit user expresses frustration over Cursor's frequent UI changes that break functionality, such as agent file search, minimize button, and MCP server connections. The post highlights a cycle of forced updates that randomize button locations and remove features, reflecting ongoing community dissatisfaction.
User reports Codex task consumed 70%+ of 5-hour allowance in 20 minutes
A Reddit user reported that a single Codex task using Extra High reasoning consumed over 70% of their 5-hour usage allowance in about 20 minutes. The task edited five files, used the browser, and ran commands. The user questions whether such high consumption is expected for Extra High reasoning tasks.
Users report Claude Code consuming excessive usage due to large context bug
Multiple users report that Claude Code consumes 20-40% of their usage allowance per session due to a bug that sends 300k+ message contexts even on fresh sessions. One user lost significant paid usage before realizing the issue. Another user describes a layered configuration setup with CLAUDE.md and AGENTS.md files, and a cleanup that removed ~700MB of empty content but still faced model selection inefficiencies.
Users complain about pinned prompt in Claude Code overlay update on Cursor
After a recent Claude Code overlay update in Cursor, the user's prompt is now permanently pinned to the top while Claude generates a response, reducing the visible space for streaming output. Users are seeking ways to revert to the previous unpinned behavior.
User reports tool-selection accuracy drops linearly with more MCP servers due to token bloat
A user connected 4 MCP servers to one agent and observed tool-selection accuracy declining linearly with server count. They traced the issue to every tool's name, description, and JSON Schema being serialized into every request, causing token bloat. With 4 servers and ~60 tools, the serialized definitions consumed significant context, degrading performance.