{
  "markdown": "# techne\n\n> **τέχνη** · /ˈtɛk.niː/ · \"TEK-nee\"\n\nAncient Greek for **craft · skill · art**: the practical know-how of *making*\nand *doing*, and the shared root of both **technique** and **technology**.\nAristotle placed *technē* (knowing **how**) beside *epistēmē* (knowing\n**that**).\n\nSkills and agents for the AI era.\n\nWebsite: [techne.wangyun.fan](https://techne.wangyun.fan).\n\nChinese version: [README-CN.md](README-CN.md).\n\n## What This Is\n\ntechne is a curated library of skills and agent assets for practical AI-era\nwork. A techne skill is not a prompt snippet. It is a forced cognitive move: a\nsmall procedure that makes an AI collaborator investigate, judge, write, code,\nor ship in a way it often skips under pressure.\n\nThe project keeps one shared source of truth in `skills/`. Host-specific files\nstay thin: Claude Code plugin metadata, Cursor and Gemini manifests, and install\ninstructions for Codex, Kimi, and other Agent Skills hosts all point back to the\nsame skill bodies.\n\n## Skills\n\nThe current seeded skills use the `anchor-*` prefix because each one anchors AI\nwork to evidence it cannot self-report or fake: a source tree, run ledger, git\ndiff, or written brief.\n\n- `anchor-viz` — investigate a real codebase by diagramming it. It routes one user\n  request across architecture, interaction, data-model, state-model, and\n  type-structure diagrams, then validates and builds a local viewer. See\n  [skills/anchor-viz/README.md](skills/anchor-viz/README.md) for detailed usage.\n- `anchor-repro` — force bug fixes through a failing reproduction first, then verify\n  with the same probe after the fix. See\n  [skills/anchor-repro/README.md](skills/anchor-repro/README.md) for detailed usage.\n- `anchor-vet` — force code review through computed scope, accounted blast radius,\n  verified claims/findings, and admissible verdicts before approving or\n  requesting changes. See [skills/anchor-vet/README.md](skills/anchor-vet/README.md) for\n  detailed usage.\n- `anchor-intake` — interrogate a written engineering implementation brief before work\n  starts, accounting the fixed rubric, surfacing gaps/traps/questions, and\n  emitting an intent-level plan. See\n  [skills/anchor-intake/README.md](skills/anchor-intake/README.md) for detailed usage.\n\n## Install\n\n**Recommended: ask your AI agent.** Paste this into Claude Code or Codex:\n\n```text\nInstall techne for my current agent by following https://github.com/lynxlangya/techne/blob/main/INSTALL.md, then verify the techne skills are available.\n```\n\nThe agent reads [INSTALL.md](INSTALL.md), detects your harness (Claude / Codex /\nCursor / Gemini / Kimi), runs the right install path, and confirms the skill\nworks.\n\n**Manual:** per-harness commands and the universal `npx skills` fallback are in\n[INSTALL.md](INSTALL.md).\n\nFor Claude Code:\n\n```text\n/plugin marketplace add lynxlangya/techne\n/plugin install techne@techne-dev\n/reload-plugins\n```\n\nFor Codex:\n\n```bash\nnpx skills add lynxlangya/techne -a codex -g -y\n```\n\nSee [WORKFLOW.md](WORKFLOW.md) for the delivery process and\n[ROADMAP.md](ROADMAP.md) for the product map.\n\n## Documentation\n\nPublic-facing documentation is split by language:\n\n- `README.md` is the default English document.\n- `README-CN.md` is the Chinese companion document.\n\nThe same convention applies to skill usage guides when a Chinese version is\nneeded, for example `skills/anchor-viz/README.md` and `skills/anchor-viz/README-CN.md`.\n",
  "bytes": 3415,
  "sha": "3c06939472546ed282c2ceb098fca3c62e86da019b99d772eb3d1b5006fd64f4",
  "repo_slug": "lynxlangya/techne",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_lynxlangya_techne_fbe6fd09/readme"
}