{"slug": "context-mode", "name": "context-mode", "category": "ai-coding", "category_name": "AI coding tools", "description": "", "recommenders": 3, "commenters": 8, "warners": 0, "score": 3.904, "rising": -0.98, "first": "2026-03-01", "latest": "2026-09-23", "why": "I prefer it over rtk / headroom (at least in Cursor).", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 3, "neutral": 4, "compare": 1, "warn": 0}, "spark": [0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0], "listed": true, "loved": [], "caveat": "", "short_quote": "I prefer it over rtk / headroom (at least in Cursor).", "kind": "", "icon": true, "domain": null, "github_repo": "github.com/mksglu/context-mode", "official_url": "https://github.com/mksglu/context-mode", "pricing": "", "aliases": ["context-mode", "Context Mode"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 771294740, "item_id": 48205921, "url": "https://news.ycombinator.com/item?id=48205921", "by": "kurtextrem", "date": "2026-05-20", "stance": "recommend", "quote": "I prefer it over rtk / headroom (at least in Cursor).", "excerpt": "Have you tried https://github.com/mksglu/context-mode? I prefer it over rtk / headroom (at least in Cursor). The stop hooks seem to work well / prevent the agent from blowing up its context window.", "context": "", "use_case": "Context management for agents", "switched_from": "", "story": "Show HN: Semble \u2013 Code search for agents that uses 98% fewer tokens than grep"}, {"id": 764709618, "item_id": 47794351, "url": "https://news.ycombinator.com/item?id=47794351", "by": "piotraleksander", "date": "2026-04-16", "stance": "recommend", "quote": "last30days and context-mode being the most useful so far", "excerpt": "I've setup one on Hetzner. I got it a dedicated email account (via agentmail.to). Installed couple of skills, last30days and context-mode being the most useful so far (https://github.com/mvanhorn/last30days-skill and https://github.com/mksglu/context-mode). The biggest unlock was wiring it up to my agency's whatsapp group - it's now basically a multiplayer chatgpt. Not a \"personal\" assistant, but more of agency's ass", "context": "", "use_case": "", "switched_from": "", "story": "Is anyone actually using OpenClaw?"}, {"id": 755296352, "item_id": 47206022, "url": "https://news.ycombinator.com/item?id=47206022", "by": "vexorkai", "date": "2026-03-01", "stance": "recommend", "quote": "Context Mode solves output compression really well", "excerpt": "This post made me realize I had zero visibility into where my Claude Code tokens were actually going, so I built a small companion CLI this morning: https://github.com/vexorkai/claude-trace It parses ~/.claude/projects/*/*.jsonl and breaks usage down by session, tool, project, and timeline with cost estimates (including cache read/create split). Context Mode solves output compression really well; this is more of a me", "context": "", "use_case": "MCP output compression", "switched_from": "", "story": "Stop Burning Your Context Window \u2013 How We Cut MCP Output by 98% in Claude Code"}]}], "caveats": [], "who": [], "tags": [], "use_cases": ["pi.dev package", "context management for agents", "plugin", "mcp output compression"], "compared": [{"slug": "rtk", "name": "RTK", "co": 2, "switched_from": 0, "switched_to": 0}, {"slug": "pi", "name": "Pi", "co": 2, "switched_from": 0, "switched_to": 0}, {"slug": "headroom", "name": "Headroom", "co": 2, "switched_from": 0, "switched_to": 0}], "makers": [{"id": 755095904, "item_id": 47193494, "url": "https://news.ycombinator.com/item?id=47193494", "by": "mksglu", "date": "2026-02-28", "stance": "recommend", "quote": "The tradeoff is intentional: allowlist over full process.env so the sandbox doesn't leak unrelated vars", "excerpt": "No magic \u2014 standard Unix process inheritance. Each execute() spawns a child process via Node's child_process.spawn() with a curated env built by #buildSafeEnv (https://github.com/mksglu/claude-context-mode/blob/main/context-mode/src/executor.ts#L245-L299). It passes through an explicit allowlist of auth vars (GH_TOKEN, AWS_ACCESS_KEY_ID, GOOGLE_APPLICATION_CREDENTIALS, KUBECONFIG, etc.) plus HOME and XDG paths so CLI", "context": "", "use_case": "", "switched_from": "", "story": "Stop Burning Your Context Window \u2013 How We Cut MCP Output by 98% in Claude Code"}, {"id": 755089184, "item_id": 47193074, "url": "https://news.ycombinator.com/item?id=47193074", "by": "mksglu", "date": "2026-02-28", "stance": "recommend", "quote": "Context Mode spawns isolated subprocesses \u2014 only stdout enters context.", "excerpt": "Author here. I shared the GitHub repo a few days ago (https://news.ycombinator.com/item?id=47148025) and got great feedback. This is the writeup explaining the architecture. The core idea: every MCP tool call dumps raw data into your 200K context window. Context Mode spawns isolated subprocesses \u2014 only stdout enters context. No LLM calls, purely algorithmic: SQLite FTS5 with BM25 ranking and Porter stemming. Since th", "context": "", "use_case": "cutting MCP tool output", "switched_from": "", "story": "Stop Burning Your Context Window \u2013 How We Cut MCP Output by 98% in Claude Code"}], "threads": [{"url": "https://news.ycombinator.com/item?id=47193064", "title": "Stop Burning Your Context Window \u2013 How We Cut MCP Output by 98% in Claude Code", "mentions": 3}, {"url": "https://news.ycombinator.com/item?id=49815568", "title": "Ask HN: Which harness do you use with Qwen3.8 27B? And why?", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48169874", "title": "Show HN: Semble \u2013 Code search for agents that uses 98% fewer tokens than grep", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48075645", "title": "The context window has been shattered: Subquadratic debuts a 12M token window", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=47994208", "title": "Context Mode v1.0.0", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=47816960", "title": "Opus 4.7 to 4.6 Inflation is ~45%", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {}, "2026-01": {}, "2026-02": {}, "2026-03": {"recommend": 1, "neutral": 1}, "2026-04": {"recommend": 1, "neutral": 1}, "2026-05": {"compare": 1, "neutral": 1, "recommend": 1}, "2026-06": {}, "2026-07": {}, "2026-08": {}, "2026-09": {"neutral": 1}, "2026-10": {}}, "quotes": [{"id": 797070609, "item_id": 49816913, "url": "https://news.ycombinator.com/item?id=49816913", "by": "slipwalker", "date": "2026-09-23", "stance": "neutral", "quote": "\"npm:context-mode\"", "excerpt": "pi.dev with \"packages\": [ \"npm:context-mode\", \"npm:pi-subagents\", \"npm:pi-web-access\", \"npm:pi-bar\", \"npm:pi-opencode-bridge\", \"npm:@ssweens/pi-leash\", \"npm:pi-caveman\", \"npm:@ff-labs/pi-fff\", \"npm:pi-ui-hephaestus\", \"npm:@alexanderfortin/pi-zai-usage\", \"npm:pi-cache-optimizer\", \"npm:@hypabolic/pi-hypa\" ]", "context": "", "use_case": "pi.dev package", "switched_from": "", "story": "Ask HN: Which harness do you use with Qwen3.8 27B? And why?"}, {"id": 771294740, "item_id": 48205921, "url": "https://news.ycombinator.com/item?id=48205921", "by": "kurtextrem", "date": "2026-05-20", "stance": "recommend", "quote": "I prefer it over rtk / headroom (at least in Cursor).", "excerpt": "Have you tried https://github.com/mksglu/context-mode? I prefer it over rtk / headroom (at least in Cursor). The stop hooks seem to work well / prevent the agent from blowing up its context window.", "context": "", "use_case": "Context management for agents", "switched_from": "", "story": "Show HN: Semble \u2013 Code search for agents that uses 98% fewer tokens than grep"}, {"id": 769332368, "item_id": 48083273, "url": "https://news.ycombinator.com/item?id=48083273", "by": "vagrantJin", "date": "2026-05-10", "stance": "neutral", "quote": "I use a tool called context-mode.", "excerpt": "I use a tool called context-mode. Update agent to save session summary every 100k tokens", "context": "", "use_case": "", "switched_from": "", "story": "The context window has been shattered: Subquadratic debuts a 12M token window"}, {"id": 767908592, "item_id": 47994287, "url": "https://news.ycombinator.com/item?id=47994287", "by": "esperent", "date": "2026-05-03", "stance": "compare", "quote": "Although this does perhaps go further than rtk/v8.", "excerpt": "Since this site seems to have an allergy to information density and usability, here's the GH: https://github.com/mksglu/context-mode Some discussion about this from a couple of months ago: https://news.ycombinator.com/item?id=47193064 Similar tools in this space: RTK:CLI proxy that reduces LLM token consumption: https://github.com/rtk-ai/rtk 8v: One CLI for you and your AI agent. Posted here a few days ago: https://n", "context": "", "use_case": "", "switched_from": "", "story": "Context Mode v1.0.0"}, {"id": 765094080, "item_id": 47818380, "url": "https://news.ycombinator.com/item?id=47818380", "by": "vidarh", "date": "2026-04-18", "stance": "neutral", "quote": "I assume they mean", "excerpt": "I assume they mean: https://github.com/mksglu/context-mode", "context": "", "use_case": "Plugin", "switched_from": "", "story": "Opus 4.7 to 4.6 Inflation is ~45%"}, {"id": 764709618, "item_id": 47794351, "url": "https://news.ycombinator.com/item?id=47794351", "by": "piotraleksander", "date": "2026-04-16", "stance": "recommend", "quote": "last30days and context-mode being the most useful so far", "excerpt": "I've setup one on Hetzner. I got it a dedicated email account (via agentmail.to). Installed couple of skills, last30days and context-mode being the most useful so far (https://github.com/mvanhorn/last30days-skill and https://github.com/mksglu/context-mode). The biggest unlock was wiring it up to my agency's whatsapp group - it's now basically a multiplayer chatgpt. Not a \"personal\" assistant, but more of agency's ass", "context": "", "use_case": "", "switched_from": "", "story": "Is anyone actually using OpenClaw?"}, {"id": 756843489, "item_id": 47302718, "url": "https://news.ycombinator.com/item?id=47302718", "by": "atombender", "date": "2026-03-08", "stance": "neutral", "quote": "I was using this plugin [1] for a bit", "excerpt": "That is very useful. I wasn't sure if I could supply my own override list or how I would even format one, but this solves that problem! The process control policy, that's kind of niche and should definitely not be something agents are always allowed to do, so having a shorthand flag like you added in that pull request is the right choice. I'm sure Anthropic and the other major players will catch up and add better san", "context": "", "use_case": "", "switched_from": "", "story": "Agent Safehouse \u2013 macOS-native sandboxing for local agents"}, {"id": 755296352, "item_id": 47206022, "url": "https://news.ycombinator.com/item?id=47206022", "by": "vexorkai", "date": "2026-03-01", "stance": "recommend", "quote": "Context Mode solves output compression really well", "excerpt": "This post made me realize I had zero visibility into where my Claude Code tokens were actually going, so I built a small companion CLI this morning: https://github.com/vexorkai/claude-trace It parses ~/.claude/projects/*/*.jsonl and breaks usage down by session, tool, project, and timeline with cost estimates (including cache read/create split). Context Mode solves output compression really well; this is more of a me", "context": "", "use_case": "MCP output compression", "switched_from": "", "story": "Stop Burning Your Context Window \u2013 How We Cut MCP Output by 98% in Claude Code"}]}