{
  "markdown": "<div align=\"center\">\n  <img src=\"docs/assets/tldr-mascot.png\" width=\"120\" alt=\"TLDR mascot\" />\n</div>\n\n<h1 align=\"center\">TLDR</h1>\n\n<p align=\"center\">\n  <strong>Verdict first. Filler never.</strong>\n</p>\n\n<p align=\"center\">\n  <a href=\"https://github.com/0p9b/TLDR/stargazers\"><img src=\"https://img.shields.io/github/stars/0p9b/TLDR?style=flat&color=yellow\" alt=\"Stars\"></a>\n  <a href=\"https://github.com/0p9b/TLDR/commits/main\"><img src=\"https://img.shields.io/github/last-commit/0p9b/TLDR?style=flat\" alt=\"Last Commit\"></a>\n  <a href=\"LICENSE\"><img src=\"https://img.shields.io/github/license/0p9b/TLDR?style=flat\" alt=\"License\"></a>\n  <img src=\"https://img.shields.io/badge/works%20with-30%2B%20agents-2c2a26?style=flat\" alt=\"30+ agents\" />\n</p>\n\n<p align=\"center\">\n  <a href=\"#before--after\">Before/After</a> •\n  <a href=\"#the-prompt--tldrmd\">Prompt</a> •\n  <a href=\"#install--prompt-only-installsh\">Prompt install</a> •\n  <a href=\"#install--full-stack-bininstalljs\">Full install</a> •\n  <a href=\"./docs/INSTALL.md\">Install</a> •\n  <a href=\"data/benchmarks.md\">Benchmarks</a>\n</p>\n\n---\n\nTerse, high-signal responses for AI coding agents — less filler, same accuracy on tools, code, and safety.\n\n**One repo, two install paths:**\n\n| Path | What you get | Best for |\n|------|----------------|----------|\n| **[`TLDR.md`](TLDR.md) + [`install.sh`](install.sh)** | Copies the system prompt (and `/tldr` command) into standard agent config locations | Quick global rules file, no Node required |\n| **[`bin/install.js`](bin/install.js)** (via `npx`) | Detects installed agents; plugins, hooks, skills, optional MCP shrink | Claude Code, Cursor, Codex, Gemini, 30+ agents — see **[docs/INSTALL.md](docs/INSTALL.md)** |\n\n> Historical prompt variants and benchmarks: [`data/changelog.md`](data/changelog.md), [`data/progression.md`](data/progression.md).\n\n## Before / After\n\n<table>\n<tr>\n<td width=\"50%\">\n\n### 🗣️ Normal agent (verbose)\n\n> \"Sure! I'd be happy to help. The reason your React component is re-rendering is likely because you're creating a new object reference on each render. I'd recommend wrapping it in `useMemo`.\"\n\n</td>\n<td width=\"50%\">\n\n### <img src=\"docs/assets/tldr-mascot.png\" width=\"20\" alt=\"TLDR\"/> TLDR mode\n\n> New object ref each render. Inline prop = new ref = re-render. `useMemo`.\n\n</td>\n</tr>\n</table>\n\n**Same fix. ~60–75% fewer prose tokens in historical benchmarks** (earlier prompt generations; the current prompt has not been rerun) — tools, code, and safety unchanged. Writeup: [`data/benchmarks.md`](data/benchmarks.md) · honest cost/benefit: [`docs/HONEST-NUMBERS.md`](docs/HONEST-NUMBERS.md).\n\n## The prompt — `TLDR.md`\n\n[`TLDR.md`](TLDR.md) is the active prompt (2,218 bytes). It changes **prose style only** — not tools, reasoning, or safety.\n\n| File | Bytes |\n|------|------:|\n| [`TLDR.md`](TLDR.md) | 2,218 |\n| [`commands/tldr.md`](commands/tldr.md) | 1,508 |\n\n**Current defaults (prompt):**\n- default: 1 sentence\n- target: 3 words\n- default max: 6 words\n- one-word greeting for plain greetings\n- `/tldr` (supported agents) re-applies rules live in long sessions\n- TLDR footer: replies to the user only that exceed 3 sentences, 30 words, or 100 characters end with a brief `TLDR:` summary line\n\n## Install — prompt only (`install.sh`)\n\nNo Node required. Writes `TLDR.md` to seven standard agent paths and installs `/tldr` where supported.\n\n```bash\ncurl -fsSL https://raw.githubusercontent.com/0p9b/TLDR/main/install.sh | bash -s --\n```\n\nOptional Hermes merge into `~/.hermes/SOUL.md`:\n\n```bash\ncurl -fsSL https://raw.githubusercontent.com/0p9b/TLDR/main/install.sh | bash -s -- --with-hermes\n```\n\n(Prompt-only. Full Hermes skill suite → `node bin/install.js --only hermes`.)\n\nPreview: `curl -fsSL https://raw.githubusercontent.com/0p9b/TLDR/main/install.sh`\n\nManual paths and copy/paste commands: [`data/agent-locations.md`](data/agent-locations.md).\n\n## Install — full stack (`bin/install.js`)\n\nNode ≥18. Auto-detects agents, installs the right plugin/extension/skills/hooks per agent.\n\n```bash\nnpx -y github:0p9b/TLDR\n```\n\nFrom a clone:\n\n```bash\ngit clone https://github.com/0p9b/TLDR.git && cd TLDR\nnode bin/install.js          # detected agents\nnode bin/install.js --all    # hooks + per-repo init (mcp-shrink needs upstream — opt in)\nnode bin/install.js --with-mcp-shrink=\"npx @modelcontextprotocol/server-filesystem /tmp\"\nnode bin/install.js --list   # agent matrix\n```\n\nWindows: [`install.ps1`](install.ps1) runs the Node installer.\n\n**Full flags, per-agent table, verify, uninstall, troubleshooting:** **[docs/INSTALL.md](docs/INSTALL.md)**.\n\n### Updating\n\n```bash\ntldr update                 # or: node bin/install.js update\ntldr update --check         # report available update only\n```\n\nRefreshes the git tree (local clone or `~/.tldr/src`), then reinstalls agent hooks/skills. Details: **[docs/INSTALL.md § Updating](docs/INSTALL.md#updating)**. Slash: `/tldr-update`.\n\n## Verify (prompt install)\n\n```bash\nfor p in ~/.claude/CLAUDE.md ~/.gemini/AGENTS.md ~/.codex/AGENTS.md \\\n         ~/AGENTS.md ~/.config/opencode/AGENTS.md \\\n         ~/.factory/AGENTS.md ~/.pi/agent/AGENTS.md; do\n  [ -f \"$p\" ] && grep -q \"^## Prime directive\" \"$p\" && echo \"✓ $p\" || echo \"✗ $p\"\ndone\ngrep -q \"^## Prime directive\" ~/.hermes/SOUL.md 2>/dev/null && echo \"✓ ~/.hermes/SOUL.md\" || echo \"✗ ~/.hermes/SOUL.md\"\n```\n\n## Repository map\n\n| Path | Purpose |\n|------|---------|\n| `TLDR.md` | Canonical terse system prompt |\n| `commands/tldr.md` | `/tldr` slash command |\n| `install.sh` | Prompt-only installer (+ optional Hermes) |\n| `install.ps1` | Windows entry → `bin/install.js` |\n| `bin/install.js` | Unified multi-agent installer |\n| `skills/` | TLDR skill suite (source of truth for behavior) |\n| `plugins/tldr/` | Claude Code / distribution mirrors (CI-synced) |\n| `src/hooks/` | Claude Code SessionStart / mode hooks |\n| `docs/INSTALL.md` | Full installer documentation |\n| `.github/CONTRIBUTING.md` | How to change skills or add agents |\n| `data/agent-locations.md` | Per-agent install paths |\n\n## Research & benchmarks\n\n- [data/agent-locations.md](data/agent-locations.md)\n- [data/benchmarks.md](data/benchmarks.md)\n- [data/dspy-cross-model-results.md](data/dspy-cross-model-results.md)\n- [data/changelog.md](data/changelog.md)\n\n## License\n\nMIT — see [`LICENSE`](LICENSE). Third-party lineage/notices: [`ATTRIBUTION.md`](docs/legal/ATTRIBUTION.md), [`THIRD_PARTY_NOTICES.md`](docs/legal/THIRD_PARTY_NOTICES.md).",
  "bytes": 6443,
  "sha": "cea29f8231b2c00dfbf67bc955a2037a778e735fa611f377be1143b564d8c800",
  "repo_slug": "0p9b/tldr",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_0p9b_tldr_ea50e4ae/readme"
}