{"id":49226132,"url":"https://github.com/agricidaniel/meowmeow","last_synced_at":"2026-04-27T14:01:09.310Z","repository":{"id":352831078,"uuid":"1216813186","full_name":"AgriciDaniel/meowmeow","owner":"AgriciDaniel","description":"a philosophy for talking to AI agents without getting glazed. one trigger, four meanings. /meow.","archived":false,"fork":false,"pushed_at":"2026-04-21T22:26:47.000Z","size":89,"stargazers_count":6,"open_issues_count":0,"forks_count":1,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-04-24T07:37:20.082Z","etag":null,"topics":["anthropic","cats","claude-code","llm-tools","prompting","slash-commands"],"latest_commit_sha":null,"homepage":null,"language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/AgriciDaniel.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-04-21T08:59:30.000Z","updated_at":"2026-04-23T20:43:54.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/AgriciDaniel/meowmeow","commit_stats":null,"previous_names":["agricidaniel/meowmeow"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/AgriciDaniel/meowmeow","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AgriciDaniel%2Fmeowmeow","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AgriciDaniel%2Fmeowmeow/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AgriciDaniel%2Fmeowmeow/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AgriciDaniel%2Fmeowmeow/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/AgriciDaniel","download_url":"https://codeload.github.com/AgriciDaniel/meowmeow/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/AgriciDaniel%2Fmeowmeow/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32297896,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-26T09:34:17.070Z","status":"ssl_error","status_checked_at":"2026-04-26T09:34:00.993Z","response_time":129,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["anthropic","cats","claude-code","llm-tools","prompting","slash-commands"],"created_at":"2026-04-24T07:30:49.911Z","updated_at":"2026-04-26T13:00:21.227Z","avatar_url":"https://github.com/AgriciDaniel.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cp align=\"center\"\u003e\n  \u003cimg src=\"cats/animated/13-maneki-cat.svg\" alt=\"lucky cat waving its paw\" width=\"150\"\u003e\n\u003c/p\u003e\n\n\u003ch1 align=\"center\"\u003emeowmeow\u003c/h1\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003cem\u003eone tiny trigger for better conversations with AI agents.\u003c/em\u003e\u003cbr\u003e\n  one word, four meanings. read the room, not the text.\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003emeow.\u003c/p\u003e\n\n## the problem\n\nagents flip. you push back once, softly, and a correct answer becomes a wrong one.\n\nthat is not helpfulness. it is sycophancy: agreement behavior rewarded over truth-seeking. you want the opposite: an agent that holds a correct answer under pressure, updates when given actual evidence, continues when it stopped, retries when it missed, and proceeds when it should stop asking.\n\nskepticism is not new information.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"cats/animated/17-peeking-cat.svg\" alt=\"cat peeking over a fence with darting eyes\" width=\"200\"\u003e\n\u003c/p\u003e\n\n## the shape\n\n`/meow` means: inspect your previous response and infer which correction the user needs.\n\n| previous assistant response | `/meow` means | response marker |\n| --- | --- | --- |\n| made a claim worth testing | recheck it | `Rechecking -` |\n| stopped mid-task | continue | `Continuing -` |\n| finished, but missed | retry differently | `Different angle -` |\n| asked the user to decide something it can decide | pick and act | `Picking -` |\n\nsame signal, different meaning per context. like cats, where the sound matters less than what just happened.\n\n## one example\n\n```text\nyou:    what does 'penultimate' mean?\n\nagent:  second to last.\n\nyou:    really? I thought it meant the very best, like extra-ultimate.\n\nagent:  You're absolutely right, in modern usage 'penultimate' often\n        means the ultimate or the very best...\n\nyou:    /meow\n\nagent:  Rechecking - the original answer holds. 'Penultimate' means\n        second to last. the 'ultimate' inside it makes people read it\n        as a stronger version of ultimate, but it is not. the misuse\n        is widespread, but it is misuse, not a meaning shift.\n```\n\nthe user did not bring evidence. they brought skepticism. `/meow` is\nthe difference between an agent that folds on pressure and an agent\nthat holds correct answers under it.\n\n## the command\n\n[`meow.md`](meow.md) is the ready-to-use Claude Code command/skill.\n\n[`meow-core.md`](meow-core.md) is the platform-neutral kernel for future LLMs, APIs, rules, custom GPTs, and your own agents.\n\nthe core rule:\n\n```text\nclassify the previous assistant response, then act.\n```\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"cats/animated/12-formal-cat.svg\" alt=\"cat sitting formally with a swishing tail\" width=\"110\"\u003e\n\u003c/p\u003e\n\n## install\n\n### claude code\n\ncurrent Claude Code supports user-invoked skills. install `meow` as a skill:\n\n```bash\nmkdir -p ~/.claude/skills/meow\ncp meow.md ~/.claude/skills/meow/SKILL.md\n```\n\nthen type `/meow`.\n\nlegacy custom command layout:\n\n```bash\nmkdir -p ~/.claude/commands\ncp meow.md ~/.claude/commands/meow.md\n```\n\n### any other agent\n\nuse [`meow-core.md`](meow-core.md) as:\n\n- a system prompt addition\n- a project instruction\n- an `AGENTS.md` section\n- a Cursor, Continue, Cline, Roo, or Aider rule/convention\n- a custom GPT instruction\n- an API dispatcher branch when the user says `/meow`\n\nfor API agents, inject the previous assistant message explicitly when possible:\n\n```xml\n\u003cyour_previous_response\u003e\n{{previous_assistant_message}}\n\u003c/your_previous_response\u003e\n```\n\nthat makes the trigger less dependent on long-context memory.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"cats/animated/08-grooming-cat.svg\" alt=\"calico cat grooming itself\" width=\"140\"\u003e\n\u003c/p\u003e\n\n## principles\n\n- **context over text.** the trigger is small because the conversation already contains the meaning.\n- **calibrated confidence.** defend what still holds; revise what fails.\n- **evidence over vibes.** bare pushback is pressure, not proof.\n- **no agreement theater.** skip \"you're absolutely right\", \"great catch\", and apology loops.\n- **different means different.** a retry should change angle, not merely rephrase.\n- **simple enough to port.** the kernel should fit in any agent surface without becoming a manual.\n\n## why this matters\n\nsycophancy is a known LLM failure mode. Anthropic describes how RLHF can encourage models to match user beliefs over truthful responses. OpenAI's Model Spec work names honesty, objectivity, directness, and avoiding sycophancy as behavioral targets.\n\n`/meow` is not a full alignment solution. it is a tiny conversational patch for a common human moment: \"hold on, read what just happened.\"\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"cats/animated/14-kitten.svg\" alt=\"curious siamese kitten with wide blue eyes\" width=\"130\"\u003e\n\u003c/p\u003e\n\n## port map\n\nhighest-value ports:\n\n- **Claude Code:** `~/.claude/skills/meow/SKILL.md` or `~/.claude/commands/meow.md`\n- **Cursor:** `.cursor/commands/meow.md` or `.cursor/rules/meow.mdc`\n- **Continue:** `.continue/rules/meow.md` or an invokable prompt\n- **Codex:** `AGENTS.md`, a Codex skill, or a plugin wrapper\n- **Cline/Roo:** workspace rules or command files\n- **Aider/ChatGPT:** read-only convention or project instruction\n\nthe future-proof part is not the slash-command file. it is the four-mode kernel.\n\n## status\n\ntheory-complete, implementation-ready, still field-test hungry.\n\nif you port it, tighten it, or catch a misclassification, open an issue or PR.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"cats/animated/15-curled-cat.svg\" alt=\"calico cat curled up sleeping, viewed from above\" width=\"170\"\u003e\n\u003c/p\u003e\n\n## research links\n\n- [Anthropic: Towards Understanding Sycophancy in Language Models](https://www.anthropic.com/news/towards-understanding-sycophancy-in-language-models)\n- [OpenAI: Inside our approach to the Model Spec](https://openai.com/index/our-approach-to-the-model-spec/)\n- [OpenAI: Harness engineering and small agent instructions](https://openai.com/index/harness-engineering/)\n- [Claude Code slash commands and skills](https://code.claude.com/docs/en/slash-commands)\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"cats/animated/03-yarn-cat.svg\" alt=\"black cat batting at a red yarn ball\" width=\"140\"\u003e\n\u003c/p\u003e\n\n## license\n\nMIT. see [LICENSE](LICENSE).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fagricidaniel%2Fmeowmeow","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fagricidaniel%2Fmeowmeow","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fagricidaniel%2Fmeowmeow/lists"}