[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7C9YGVMNK5XRFX596ZGSGJ4":3,"skill-same:01M7C9YGVMNK5XRFX596ZGSGJ4":65},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":31,"commitSha":32,"skillDirectory":33,"forksCount":14,"repositoryHtmlUrl":34,"repositoryHomepageUrl":35,"ownerHtmlUrl":36,"ownerBlogUrl":37,"ownerDescription":38,"updatedAt":39,"install":40,"resources":52},"01M7C9YGVMNK5XRFX596ZGSGJ4","install-codex-agents",null,"Install, refresh, or check crux's Codex role agents in the personal or specified project directory. Preserve conflicting roles.","bionic-coding\u002Fcrux","bionic-coding","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F301040008?v=4","01M7C9T2NA1V1V0EF1M552HRB3","Bionic Coding","01M7C9T2P72H1AQV5T5PTBFNHN",0,"MIT","https:\u002F\u002Fgithub.com\u002Fbionic-coding\u002Fcrux\u002Fblob\u002Fmain\u002Fcrux\u002Fskills\u002Finstall-codex-agents\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: install-codex-agents\ndescription: \"Install, refresh, or check crux's Codex role agents in the personal or specified project directory. Preserve conflicting roles.\"\ndisable-model-invocation: true\nmetadata:\n  tags: \"codex, agents, installation, roles\"\n  bundles: \"crux-infrastructure, crux-docs\"\n  risk_level: \"medium\"\n  requires_access: \"write-host\"\n  triggers: \"install codex agents | install crux agents for codex | refresh the codex agents\"\n---\n\n# Install Codex Agents\n\n\u003C!-- BEGIN GENERATED: runtime-compat -->\n## Runtime compatibility\n\nThis skill is portable across Claude Code, Codex, OpenCode, and OMP (Oh My Pi). This section overrides platform-specific labels below.\n\n- Before running a command that uses `CRUX_PLUGIN_ROOT`, set it to the installed plugin root. In Claude Code, use the value of `CLAUDE_PLUGIN_ROOT`. In Codex, OpenCode, and OMP, derive it from the absolute path of this selected `SKILL.md`: the plugin root is the parent of its `skills\u002F` directory (in OMP `bash`, `skill:\u002F\u002F\u003Cname>` resolves to the skill directory). In a source checkout, use the checkout `crux\u002F` directory.\n- For project-local skills, use `.claude\u002Fskills` in Claude Code, `.agents\u002Fskills` in Codex, `.opencode\u002Fskills` in OpenCode (which also reads the singular `.opencode\u002Fskill`), and `.omp\u002Fskills` in OMP. Set `CRUX_LOCAL_SKILLS_DIR` to that path before following any command below that uses it.\n- Translate Claude Code tool labels such as `Agent`, `Read`, `Write`, `Bash`, `WebSearch`, and `WebFetch` to the matching capability in the current session. Codex names its own capabilities; OpenCode uses the lowercase forms `subagent`, `read`, `edit`, `shell`, `websearch`, and `webfetch`, where `edit` covers both `Edit` and `Write`; OMP uses `task`, `read`, `edit`, `write`, `bash`, `grep`, `glob`, `find`, `todo`, and `web_search` (where `read` also handles URLs and `skill:\u002F\u002F` URIs). Do not attempt to invoke the Claude Code labels as literal commands on another host.\n- Install the generated role agents before delegating: `install-codex-agents` in Codex, `install-opencode-agents` in OpenCode, `install-omp-agents` in OMP. Codex names them `crux_architect`, `crux_brainstormer`, `crux_commander`, `crux_dev_lead`, `crux_developer`, `crux_historian`, `crux_librarian`, `crux_night_gardener`, `crux_reviewer`, and `crux_wayfinder`; OpenCode and OMP use the bare role names `architect`, `brainstormer`, `commander`, `dev-lead`, `developer`, `historian`, `librarian`, `night-gardener`, `reviewer`, and `wayfinder` (in OMP, `scout` may also be used for reconnaissance alongside `wayfinder`). If a required role or capability is unavailable, report that truthfully instead of claiming it ran.\n- Argument placeholders such as `$adr` and `$book` bind only in Claude Code. On a host without argument binding they are unset — take the value from the user's phrase. The \"Fields OpenCode ignores\" section of `OPENCODE_GUIDE.md` names the invocation-control fields OpenCode ignores.\n\u003C!-- END GENERATED: runtime-compat -->\n\n\n## Overview\n\nCodex reads custom subagents from `.codex\u002Fagents\u002F*.toml`; it does not consume\nCrux's source agent Markdown files. This skill installs the ten namespaced Crux\nroles into the active user's personal Codex directory by default:\n\n`crux_architect`, `crux_brainstormer`, `crux_commander`, `crux_dev_lead`,\n`crux_developer`, `crux_historian`, `crux_librarian`,\n`crux_night_gardener`, `crux_reviewer`, and `crux_wayfinder`.\n\nThe generated files are `crux-\u003Crole>.toml`. Each file pins the role's model and\nreasoning effort from `crux\u002Fcatalog\u002Fmodels.yml`. It also contains one enabled\n`skills.config` entry for each skill declared in `crux\u002Fagents\u002F\u003Crole>.md`.\nThose entries use absolute paths to the selected plugin's `SKILL.md` files.\n\nThe installer preserves unrelated agent files. Read-oriented roles default to\nCodex's `read-only` sandbox; writer roles default to `workspace-write`.\n\n## Install\n\n\u003C!-- Provenance: this skill assumes the crux plugin is already installed in\nCodex via `codex plugin marketplace add bionic-coding\u002Fcrux` then `codex plugin add\ncrux@crux` (documented in install-docs-skills\u002Freferences\u002Finstall-and-upgrade.md). That CLI syntax is\nan external Codex contract owned by OpenAI, not by crux — track it against\nupstream, don't assume it. Verified 2026-07-09 against the official docs\n(https:\u002F\u002Fdevelopers.openai.com\u002Fcodex\u002Fplugins\u002Fbuild) and against the local\n`codex plugin ... --help` for codex-cli 0.144.0, which accept the\n`owner\u002Frepo` marketplace source and the `PLUGIN@MARKETPLACE` positional\n(`crux@crux`) forms. -->\n\nCodex provides this selected skill's absolute `SKILL.md` path in context. Let\n`\u003Cskill-dir>` be its containing directory.\n\nInstall for the active user:\n\n```bash\nuv run \u003Cskill-dir>\u002Fscripts\u002Finstall.py\n```\n\nThis targets `~\u002F.codex\u002Fagents\u002F`. Supply the current repository when you also\nwant the health report to scan for project agents with the same effective name:\n\n```bash\nuv run \u003Cskill-dir>\u002Fscripts\u002Finstall.py --project-context \"$PWD\"\n```\n\nUse project scope only when the user requests repository-local agents:\n\n```bash\nuv run \u003Cskill-dir>\u002Fscripts\u002Finstall.py --repo-root \"$PWD\"\n```\n\n`--repo-root` and `--codex-home` are mutually exclusive. `--codex-home \u003Cpath>`\nsets the personal target to `\u003Cpath>\u002Fagents`. It supports isolated tests and\nexplicit alternate installations. Codex discovery from an alternate home\nremains unverified until that exact client and path pass the fresh-session procedure in\n[`references\u002Ffresh-session-verification.md`](references\u002Ffresh-session-verification.md).\n\n### Check health without writing\n\n```bash\nuv run \u003Cskill-dir>\u002Fscripts\u002Finstall.py --check --project-context \"$PWD\"\n```\n\n`--check` prints JSON and does not create the target. Its exit status is `0`\nwhen `drift.status` and `installed.status` are `clean` and the shadow scan has\nno findings. It returns `1` for managed drift, installed-state findings, or\nshadow findings. It returns `2` for a configuration or environment error.\n\nThe stable health report uses these terms:\n\n- `scope`: `personal` or `project`.\n- `target`: the resolved agent directory.\n- `plugin`: `root` and `version` for the selected Crux plugin.\n- `roles`: the sorted canonical expectations. Each record contains `role`,\n  `name`, `model`, `model_reasoning_effort`, `declared_skills`, and\n  `resolved_skills`.\n- `installed`: parsed managed TOML state. It contains `status` (`clean` or\n  `findings`) and a sorted `roles` array. Each installed role contains `role`,\n  `file`, `status`, `model`, `model_reasoning_effort`, and `skill_bindings`.\n  Role status is `installed`, `missing`, `unsafe`, or `malformed`.\n- `skill_bindings`: each installed role's binding comparison. Every binding\n  contains `skill`, `expected_path`, `installed_path`, and `status`. Binding\n  status is `enabled`, `disabled`, or `missing`.\n- `drift`: `status` (`clean` or `drift`) plus `added`, `changed`, and `removed`.\n- `shadows`: `status` (`unverified`, `clean`, or `findings`) plus\n  `project_context`, `roles`, `malformed`, and `unsafe`.\n- `runtime`: `status`, `client_version`, and `reason`. The current runtime\n  status is `unverified`, the client version is `null`, and the reason is\n  `fresh-session host evidence has not been recorded`.\n\nA completed install adds `written` and `removed` to the report. Repeating an\nunchanged install returns an empty `written` list and does not rewrite files.\n`roles` always describes the selected plugin's canonical configuration.\n`installed.roles` describes what the health check parsed from managed files.\n\n### Refresh managed files\n\nA later run refuses to replace a changed managed file or remove a stale managed\nfile. Review the reported `changed` and `removed` lists, then refresh:\n\n```bash\nuv run \u003Cskill-dir>\u002Fscripts\u002Finstall.py --force --project-context \"$PWD\"\n```\n\nPlugin relocation changes the absolute skill paths and therefore appears as\nmanaged drift. Refresh with `--force` after you verify the new plugin root.\nThe installer preserves unrelated files and refuses managed symlink leaves.\nIt also refuses any target or managed leaf that resolves outside the selected\npersonal or project root.\n\n### Understand shadowing and runtime limits\n\n`--project-context \u003Crepo>` scans `\u003Crepo>\u002F.codex\u002Fagents\u002F*.toml` by the effective\nTOML `name`, regardless of filename. It reports matching `crux_*` roles,\nmalformed TOML, and unsafe symlinks without changing the repository. Without\n`--project-context`, `shadows.status` is `unverified`.\n\nStart a new Codex thread after installation so Codex can discover the agents.\nAsk Codex to spawn a namespaced role such as `crux_reviewer` or\n`crux_developer`. Static health separates canonical expectations in `roles`\nfrom parsed managed files in `installed.roles`. It does not prove host\ndiscovery, effective model settings, skill loading, or workflow behavior.\nRuntime stays `unverified` unless the selected Codex version has a recorded\npass and the health reporter consumes that evidence.\n\nSkill configuration controls availability and enablement. It does not grant\ntools, credentials, sandbox access, or eager instruction loading. Parent-session\npermission overrides and Codex's delegation-depth limits still apply.\n\n## Verification\n\n- [ ] The selected target contains the ten generated `crux-*.toml` files.\n- [ ] Existing non-Crux agent files remain unchanged.\n- [ ] Every role record contains the expected model, effort, and declared skills.\n- [ ] Every `resolved_skills` entry points to the selected plugin's `SKILL.md`.\n- [ ] `installed.status` is `clean` and every installed role has status\n  `installed`.\n- [ ] Installed models, efforts, and skill bindings match the canonical role.\n- [ ] `drift.status` is `clean` after installation.\n- [ ] Shadow findings were reviewed, or `shadows.status` is recorded as\n  `unverified` because no project context was supplied.\n- [ ] `runtime.status` is reported as `unverified` until fresh-session host\n  evidence exists.\n\n## Guardrails\n\n- Do not hand-edit generated `crux-*.toml` files. Change `crux\u002Fagents\u002F*.md` in\n  the Crux source, regenerate, and refresh intentionally.\n- Do not use `--force` before reviewing changed and stale managed files.\n- Do not treat a role's self-report as evidence of its effective model, effort,\n  or loaded skills. Use host-observed evidence.\n- Codex parent-session permission overrides can be broader than a role's default\n  sandbox. The role prompt remains binding even when that occurs.\n","# Install Codex Agents\n\n\u003C!-- BEGIN GENERATED: runtime-compat -->\n## Runtime compatibility\n\nThis skill is portable across Claude Code, Codex, OpenCode, and OMP (Oh My Pi). This section overrides platform-specific labels below.\n\n- Before running a command that uses `CRUX_PLUGIN_ROOT`, set it to the installed plugin root. In Claude Code, use the value of `CLAUDE_PLUGIN_ROOT`. In Codex, OpenCode, and OMP, derive it from the absolute path of this selected `SKILL.md`: the plugin root is the parent of its `skills\u002F` directory (in OMP `bash`, `skill:\u002F\u002F\u003Cname>` resolves to the skill directory). In a source checkout, use the checkout `crux\u002F` directory.\n- For project-local skills, use `.claude\u002Fskills` in Claude Code, `.agents\u002Fskills` in Codex, `.opencode\u002Fskills` in OpenCode (which also reads the singular `.opencode\u002Fskill`), and `.omp\u002Fskills` in OMP. Set `CRUX_LOCAL_SKILLS_DIR` to that path before following any command below that uses it.\n- Translate Claude Code tool labels such as `Agent`, `Read`, `Write`, `Bash`, `WebSearch`, and `WebFetch` to the matching capability in the current session. Codex names its own capabilities; OpenCode uses the lowercase forms `subagent`, `read`, `edit`, `shell`, `websearch`, and `webfetch`, where `edit` covers both `Edit` and `Write`; OMP uses `task`, `read`, `edit`, `write`, `bash`, `grep`, `glob`, `find`, `todo`, and `web_search` (where `read` also handles URLs and `skill:\u002F\u002F` URIs). Do not attempt to invoke the Claude Code labels as literal commands on another host.\n- Install the generated role agents before delegating: `install-codex-agents` in Codex, `install-opencode-agents` in OpenCode, `install-omp-agents` in OMP. Codex names them `crux_architect`, `crux_brainstormer`, `crux_commander`, `crux_dev_lead`, `crux_developer`, `crux_historian`, `crux_librarian`, `crux_night_gardener`, `crux_reviewer`, and `crux_wayfinder`; OpenCode and OMP use the bare role names `architect`, `brainstormer`, `commander`, `dev-lead`, `developer`, `historian`, `librarian`, `night-gardener`, `reviewer`, and `wayfinder` (in OMP, `scout` may also be used for reconnaissance alongside `wayfinder`). If a required role or capability is unavailable, report that truthfully instead of claiming it ran.\n- Argument placeholders such as `$adr` and `$book` bind only in Claude Code. On a host without argument binding they are unset — take the value from the user's phrase. The \"Fields OpenCode ignores\" section of `OPENCODE_GUIDE.md` names the invocation-control fields OpenCode ignores.\n\u003C!-- END GENERATED: runtime-compat -->\n\n\n## Overview\n\nCodex reads custom subagents from `.codex\u002Fagents\u002F*.toml`; it does not consume\nCrux's source agent Markdown files. This skill installs the ten namespaced Crux\nroles into the active user's personal Codex directory by default:\n\n`crux_architect`, `crux_brainstormer`, `crux_commander`, `crux_dev_lead`,\n`crux_developer`, `crux_historian`, `crux_librarian`,\n`crux_night_gardener`, `crux_reviewer`, and `crux_wayfinder`.\n\nThe generated files are `crux-\u003Crole>.toml`. Each file pins the role's model and\nreasoning effort from `crux\u002Fcatalog\u002Fmodels.yml`. It also contains one enabled\n`skills.config` entry for each skill declared in `crux\u002Fagents\u002F\u003Crole>.md`.\nThose entries use absolute paths to the selected plugin's `SKILL.md` files.\n\nThe installer preserves unrelated agent files. Read-oriented roles default to\nCodex's `read-only` sandbox; writer roles default to `workspace-write`.\n\n## Install\n\n\u003C!-- Provenance: this skill assumes the crux plugin is already installed in\nCodex via `codex plugin marketplace add bionic-coding\u002Fcrux` then `codex plugin add\ncrux@crux` (documented in install-docs-skills\u002Freferences\u002Finstall-and-upgrade.md). That CLI syntax is\nan external Codex contract owned by OpenAI, not by crux — track it against\nupstream, don't assume it. Verified 2026-07-09 against the official docs\n(https:\u002F\u002Fdevelopers.openai.com\u002Fcodex\u002Fplugins\u002Fbuild) and against the local\n`codex plugin ... --help` for codex-cli 0.144.0, which accept the\n`owner\u002Frepo` marketplace source and the `PLUGIN@MARKETPLACE` positional\n(`crux@crux`) forms. -->\n\nCodex provides this selected skill's absolute `SKILL.md` path in context. Let\n`\u003Cskill-dir>` be its containing directory.\n\nInstall for the active user:\n\n```bash\nuv run \u003Cskill-dir>\u002Fscripts\u002Finstall.py\n```\n\nThis targets `~\u002F.codex\u002Fagents\u002F`. Supply the current repository when you also\nwant the health report to scan for project agents with the same effective name:\n\n```bash\nuv run \u003Cskill-dir>\u002Fscripts\u002Finstall.py --project-context \"$PWD\"\n```\n\nUse project scope only when the user requests repository-local agents:\n\n```bash\nuv run \u003Cskill-dir>\u002Fscripts\u002Finstall.py --repo-root \"$PWD\"\n```\n\n`--repo-root` and `--codex-home` are mutually exclusive. `--codex-home \u003Cpath>`\nsets the personal target to `\u003Cpath>\u002Fagents`. It supports isolated tests and\nexplicit alternate installations. Codex discovery from an alternate home\nremains unverified until that exact client and path pass the fresh-session procedure in\n[`references\u002Ffresh-session-verification.md`](references\u002Ffresh-session-verification.md).\n\n### Check health without writing\n\n```bash\nuv run \u003Cskill-dir>\u002Fscripts\u002Finstall.py --check --project-context \"$PWD\"\n```\n\n`--check` prints JSON and does not create the target. Its exit status is `0`\nwhen `drift.status` and `installed.status` are `clean` and the shadow scan has\nno findings. It returns `1` for managed drift, installed-state findings, or\nshadow findings. It returns `2` for a configuration or environment error.\n\nThe stable health report uses these terms:\n\n- `scope`: `personal` or `project`.\n- `target`: the resolved agent directory.\n- `plugin`: `root` and `version` for the selected Crux plugin.\n- `roles`: the sorted canonical expectations. Each record contains `role`,\n  `name`, `model`, `model_reasoning_effort`, `declared_skills`, and\n  `resolved_skills`.\n- `installed`: parsed managed TOML state. It contains `status` (`clean` or\n  `findings`) and a sorted `roles` array. Each installed role contains `role`,\n  `file`, `status`, `model`, `model_reasoning_effort`, and `skill_bindings`.\n  Role status is `installed`, `missing`, `unsafe`, or `malformed`.\n- `skill_bindings`: each installed role's binding comparison. Every binding\n  contains `skill`, `expected_path`, `installed_path`, and `status`. Binding\n  status is `enabled`, `disabled`, or `missing`.\n- `drift`: `status` (`clean` or `drift`) plus `added`, `changed`, and `removed`.\n- `shadows`: `status` (`unverified`, `clean`, or `findings`) plus\n  `project_context`, `roles`, `malformed`, and `unsafe`.\n- `runtime`: `status`, `client_version`, and `reason`. The current runtime\n  status is `unverified`, the client version is `null`, and the reason is\n  `fresh-session host evidence has not been recorded`.\n\nA completed install adds `written` and `removed` to the report. Repeating an\nunchanged install returns an empty `written` list and does not rewrite files.\n`roles` always describes the selected plugin's canonical configuration.\n`installed.roles` describes what the health check parsed from managed files.\n\n### Refresh managed files\n\nA later run refuses to replace a changed managed file or remove a stale managed\nfile. Review the reported `changed` and `removed` lists, then refresh:\n\n```bash\nuv run \u003Cskill-dir>\u002Fscripts\u002Finstall.py --force --project-context \"$PWD\"\n```\n\nPlugin relocation changes the absolute skill paths and therefore appears as\nmanaged drift. Refresh with `--force` after you verify the new plugin root.\nThe installer preserves unrelated files and refuses managed symlink leaves.\nIt also refuses any target or managed leaf that resolves outside the selected\npersonal or project root.\n\n### Understand shadowing and runtime limits\n\n`--project-context \u003Crepo>` scans `\u003Crepo>\u002F.codex\u002Fagents\u002F*.toml` by the effective\nTOML `name`, regardless of filename. It reports matching `crux_*` roles,\nmalformed TOML, and unsafe symlinks without changing the repository. Without\n`--project-context`, `shadows.status` is `unverified`.\n\nStart a new Codex thread after installation so Codex can discover the agents.\nAsk Codex to spawn a namespaced role such as `crux_reviewer` or\n`crux_developer`. Static health separates canonical expectations in `roles`\nfrom parsed managed files in `installed.roles`. It does not prove host\ndiscovery, effective model settings, skill loading, or workflow behavior.\nRuntime stays `unverified` unless the selected Codex version has a recorded\npass and the health reporter consumes that evidence.\n\nSkill configuration controls availability and enablement. It does not grant\ntools, credentials, sandbox access, or eager instruction loading. Parent-session\npermission overrides and Codex's delegation-depth limits still apply.\n\n## Verification\n\n- [ ] The selected target contains the ten generated `crux-*.toml` files.\n- [ ] Existing non-Crux agent files remain unchanged.\n- [ ] Every role record contains the expected model, effort, and declared skills.\n- [ ] Every `resolved_skills` entry points to the selected plugin's `SKILL.md`.\n- [ ] `installed.status` is `clean` and every installed role has status\n  `installed`.\n- [ ] Installed models, efforts, and skill bindings match the canonical role.\n- [ ] `drift.status` is `clean` after installation.\n- [ ] Shadow findings were reviewed, or `shadows.status` is recorded as\n  `unverified` because no project context was supplied.\n- [ ] `runtime.status` is reported as `unverified` until fresh-session host\n  evidence exists.\n\n## Guardrails\n\n- Do not hand-edit generated `crux-*.toml` files. Change `crux\u002Fagents\u002F*.md` in\n  the Crux source, regenerate, and refresh intentionally.\n- Do not use `--force` before reviewing changed and stale managed files.\n- Do not treat a role's self-report as evidence of its effective model, effort,\n  or loaded skills. Use host-observed evidence.\n- Codex parent-session permission overrides can be broader than a role's default\n  sandbox. The role prompt remains binding even when that occurs.\n",{"name":5,"metadata":24,"description":7,"disable-model-invocation":30},{"tags":25,"bundles":26,"triggers":27,"risk_level":28,"requires_access":29},"codex, agents, installation, roles","crux-infrastructure, crux-docs","install codex agents | install crux agents for codex | refresh the codex agents","medium","write-host",true,"main","dd4f16e7ac12ff940790660ec897677284ae4b6d","crux\u002Fskills\u002Finstall-codex-agents","https:\u002F\u002Fgithub.com\u002Fbionic-coding\u002Fcrux","https:\u002F\u002Fbionic-coding.com\u002Fcrux\u002F","https:\u002F\u002Fgithub.com\u002Fbionic-coding","https:\u002F\u002Fbionic-coding.com","Durable intent, disposable code.","2026-11-12 09:36:33",{"source":8,"skillName":5,"commands":41,"prompt":51,"zipAvailable":30},[42,45,48],{"tool":43,"command":44},"npx","npx skills add bionic-coding\u002Fcrux --skill install-codex-agents",{"tool":46,"command":47},"bunx","bunx skills add bionic-coding\u002Fcrux --skill install-codex-agents",{"tool":49,"command":50},"pnpm","pnpm dlx skills add bionic-coding\u002Fcrux --skill install-codex-agents","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fbionic-coding\u002Fcrux\u002Fblob\u002Fmain\u002Fcrux\u002Fskills\u002Finstall-codex-agents\u002FSKILL.md",[53,59],{"path":54,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":57,"sourceHtmlUrl":58},"references\u002Ffresh-session-verification.md","REFERENCE","参考文档",5409,"https:\u002F\u002Fgithub.com\u002Fbionic-coding\u002Fcrux\u002Fblob\u002Fmain\u002Fcrux\u002Fskills\u002Finstall-codex-agents\u002Freferences\u002Ffresh-session-verification.md",{"path":60,"resourceType":61,"resourceTypeDesc":62,"sizeBytes":63,"sourceHtmlUrl":64},"scripts\u002Finstall.py","SCRIPT","脚本",5136,"https:\u002F\u002Fgithub.com\u002Fbionic-coding\u002Fcrux\u002Fblob\u002Fmain\u002Fcrux\u002Fskills\u002Finstall-codex-agents\u002Fscripts\u002Finstall.py",[66,73,80,87,94,101],{"skillNo":67,"name":68,"nameZh":6,"description":69,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":19},"01M7CA5PH4JT2V5WBTA5ZPJNS8","whiteboarding","Explore an idea through docs-informed dialogue before committing to a design. Capture the exploration for later filing as a brief.","https:\u002F\u002Fgithub.com\u002Fbionic-coding\u002Fcrux\u002Fblob\u002Fmain\u002Fcrux\u002Fskills\u002Fwhiteboarding\u002FSKILL.md",[],[],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":19},"01M7CA5GZPSVNJJZYQMF0JT3EH","verify-code-docs","Check generated code documentation against source and report drift without regenerating it.","https:\u002F\u002Fgithub.com\u002Fbionic-coding\u002Fcrux\u002Fblob\u002Fmain\u002Fcrux\u002Fskills\u002Fverify-code-docs\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":19},"01M7CA5A4HA73VF91N8VY6SDBN","transition-observation","Apply a human disposition to an observation's lifecycle state without changing its claim or body.","https:\u002F\u002Fgithub.com\u002Fbionic-coding\u002Fcrux\u002Fblob\u002Fmain\u002Fcrux\u002Fskills\u002Ftransition-observation\u002FSKILL.md",[],[],{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":19},"01M7CA53X1ESGWKTPCEQ55JWEZ","transition-invariant","Ratify, reject, or retire an invariant on human direction. Update lifecycle metadata without changing the invariant's body.","https:\u002F\u002Fgithub.com\u002Fbionic-coding\u002Fcrux\u002Fblob\u002Fmain\u002Fcrux\u002Fskills\u002Ftransition-invariant\u002FSKILL.md",[],[],{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":19},"01M7CA4Y2VC25NFZE0XJZ47VSX","transition-decision","Apply a human ratify, reject, or defer verdict to a recovered decision candidate; ratification creates a Proposed ADR or observation.","https:\u002F\u002Fgithub.com\u002Fbionic-coding\u002Fcrux\u002Fblob\u002Fmain\u002Fcrux\u002Fskills\u002Ftransition-decision\u002FSKILL.md",[],[],{"skillNo":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":19},"01M7CA4QD9RFDC379VD9DNK90D","transition-brief","Publish or abandon a draft brief, updating lifecycle metadata and indexes without changing its body.","https:\u002F\u002Fgithub.com\u002Fbionic-coding\u002Fcrux\u002Fblob\u002Fmain\u002Fcrux\u002Fskills\u002Ftransition-brief\u002FSKILL.md",[],[]]