[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-146":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":14,"subscribersCount":14,"size":14,"stars1d":14,"stars7d":15,"stars30d":16,"stars90d":14,"forks30d":14,"starsTrendScore":14,"compositeScore":17,"rankGlobal":9,"rankLanguage":9,"license":18,"archived":19,"fork":19,"defaultBranch":20,"hasWiki":21,"hasPages":19,"topics":22,"createdAt":9,"pushedAt":9,"updatedAt":23,"readmeContent":24,"aiSummary":25,"trendingCount":14,"starSnapshotCount":14,"syncStatus":26,"lastSyncTime":27,"discoverSource":28},146,"goal-forge","michaelpersonal\u002Fgoal-forge","michaelpersonal","Codex skill: turn a rough coding idea into a SPEC.md, GOAL.md, and \u002Fgoal-ready contract.",null,"Python",204,15,120,0,3,68,3.61,"MIT License",false,"main",true,[],"2026-06-12 02:00:08","# goal-forge\n\nCodex skill that turns a rough coding idea into a Codex `\u002Fgoal`-ready contract.\n\nPipeline: **rough idea → interviewed `SPEC.md` → tightened `SPEC.md` → `GOAL.md` → config readiness check**.\n\n## Modes\n\n- **Interview** — open-ended interview that forces decisions on scope, architecture, edge cases, verification. Hard gate: spec is not \"done\" until `done_when` has user-approved measurable criteria.\n- **Tighten** — read `SPEC.md` skeptically; surface ambiguities with two interpretations + a recommendation.\n- **Compile** — emit `GOAL.md` using the XML block structure in `references\u002Fgoal_prompt_blocks.md`. Weak specs route back to Interview\u002FTighten.\n- **Check config** — run `scripts\u002Finspect_codex_config.py` for a read-only report on Codex version, project trust, and the full autonomous `\u002Fgoal` config.\n\n## Install\n\nDrop the folder into your Codex skills directory:\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fmichaelpersonal\u002Fgoal-forge.git ~\u002F.codex\u002Fskills\u002Fgoal-forge\n```\n\nThen invoke from Codex with `$goal-forge` or any of the natural-language triggers in `SKILL.md`'s frontmatter.\n\n## Autonomous `\u002Fgoal` config\n\nFor multi-hour autonomous `\u002Fgoal` sessions, the skill checks `~\u002F.codex\u002Fconfig.toml` against this target configuration:\n\n```toml\nmodel = \"gpt-5.5\"\nmodel_context_window = 1050000\nmodel_auto_compact_token_limit = 997500\nmodel_reasoning_effort = \"high\"\nplan_mode_reasoning_effort = \"xhigh\"\napproval_policy = \"never\"\nsandbox_mode = \"danger-full-access\"\n\n[features]\ngoals = true\n```\n\n`model_reasoning_effort = \"high\"` is for execution work. `plan_mode_reasoning_effort = \"xhigh\"` is for the planning pass. `model_auto_compact_token_limit = 997500` lets long sessions compact context before they hit the hard context limit.\n\nUse `approval_policy = \"never\"` and `sandbox_mode = \"danger-full-access\"` only in project paths you have explicitly marked trusted. They give the agent unsupervised filesystem access.\n\nRun the inspector from a target project root:\n\n```bash\npython3 ~\u002F.codex\u002Fskills\u002Fgoal-forge\u002Fscripts\u002Finspect_codex_config.py --project-path \"$PWD\"\n```\n\nThe report prints `autonomous_goal_status: ready` only when the config, feature flag, Codex version, and project trust are aligned.\n\n## Layout\n\n```\ngoal-forge\u002F\n├── SKILL.md\n├── agents\u002Fopenai.yaml                       UI metadata + implicit invocation\n├── references\u002F\n│   ├── goal_prompt_blocks.md                GOAL.md XML structure\n│   ├── config_checklist.md                  Long-running \u002Fgoal config notes\n│   └── standard_execution_rules.md          Compile-time execution rules\n└── scripts\u002F\n    └── inspect_codex_config.py              Read-only config readiness report\n```\n\n## Credits\n\nInspired by [@ynkzlk](https:\u002F\u002Fx.com\u002Fynkzlk)'s blog post [*Codex \u002Fgoal: A Six-Hour Run*](https:\u002F\u002Fwww.tectontide.com\u002Fen\u002Fblog\u002Fcodex-goal-six-hour-run\u002F), which makes the case that long-running `\u002Fgoal` runs succeed or fail on upfront specification discipline — explicit measurable `done_when` criteria, XML-structured prompts, and context architecture (reading lists, working rules, anti-pattern fences) that keep the agent from taking shortcuts. This skill operationalizes that discipline as a repeatable pipeline.\n\n## License\n\nMIT. See `LICENSE`.\n","goal-forge 是一个将粗略的编码想法转化为详细规格文档和目标就绪合同的Codex技能。其核心功能包括通过一系列步骤（从初步构想到最终配置检查）自动生成`SPEC.md`、`GOAL.md`文件，并为Codex `\u002Fgoal`做好准备。技术特点涵盖了一个开放式的访谈模式来明确范围与架构，以及对生成的规范进行严格审查以消除歧义。此外，该项目还提供了一个脚本来检查系统配置是否适合长时间自动执行任务。适用于需要将非正式编程概念快速且准确地转化为具体实施计划的软件开发场景中使用。",2,"2026-06-11 02:31:06","CREATED_QUERY"]