[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M72V643VG4C97V3X764ZW12G":3,"skill-same:01M72V643VG4C97V3X764ZW12G":63},{"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":25,"commitSha":26,"skillDirectory":5,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":30,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M72V643VG4C97V3X764ZW12G","timesharing",null,"Multiplexes one agent's attention across independent long-running external processes while preserving one reasoning context. Use for concurrent tests, builds, servers, watchers, REPLs, benchmarks, downloads, and other latency-bearing shell work; never use it to spawn or substitute subagents.","roerohan\u002Fskills","roerohan","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F42958812?v=4","01M72V46MPJK55W3BZ44Y869D1","Rohan Mukherjee","01M72V46N9GPDNRC4RHVRR2YCB",29,"MIT","https:\u002F\u002Fgithub.com\u002Froerohan\u002Fskills\u002Fblob\u002Fmain\u002Ftimesharing\u002FSKILL.md",[],[],"2026-11-10 22:39:42",false,"---\nname: timesharing\ndescription: Multiplexes one agent's attention across independent long-running external processes while preserving one reasoning context. Use for concurrent tests, builds, servers, watchers, REPLs, benchmarks, downloads, and other latency-bearing shell work; never use it to spawn or substitute subagents.\ncompatibility: Prefers tmux; supports a non-interactive shell-background fallback on macOS and Linux.\n---\n\n# Timesharing\n\nTimesharing lets one agent preserve one coherent reasoning context while opportunistically multiplexing its attention across independent external work.\n\nThe AI's reasoning stays serial. Only ordinary OS processes run concurrently. Never spawn a subagent, delegate interpretation, or create another AI context for this workflow.\n\n## Central Rule\n\n> Never block on an independent long-running process when useful independent work exists.\n\nTreat yourself as a cooperative scheduler:\n\n1. Identify a command whose result is not immediately required.\n2. Start it asynchronously and record its purpose, provenance, dependency, and revisit point.\n3. Continue concrete independent work immediately.\n4. Revisit at a natural boundary, not through busy polling.\n5. Interpret the result in the same context that launched it.\n\nUse a small number of jobs, usually 2-4. The goal is to hide external latency without losing cognitive coherence, not to maximize concurrency.\n\n## Decide\n\nTimeshare when judgment says all of these are true:\n\n- The command may take meaningful time or must remain alive.\n- Its result is not required for the next reasoning step.\n- Useful independent work exists, or later interaction\u002Fincremental output matters.\n- Its inputs are stable enough that the run will remain useful.\n\nTypical jobs include test suites, builds, installs, Docker builds, servers, watchers, long scripts, migrations, data\u002Fmodel processing, downloads, benchmarks, integration tests, REPLs, and debuggers.\n\nRun trivial or immediately blocking commands normally: `pwd`, `git status`, small reads, quick searches, and commands whose answer determines the very next step. Use judgment, not a hard timing threshold.\n\n## Schedule\n\nBefore launch, form a tiny dependency graph. Only overlap genuinely independent work. Do not start a run whose inputs are likely to be invalidated immediately.\n\nMaintain a lightweight in-context ledger:\n\n```text\nJOB       PURPOSE             LOCATION                 STATE    DEPENDS\u002FREVISIT\nts-tests  validate parser     tmux:ts-repo-tests-...   running  after serializer edit\nts-api    integration target  tmux:ts-repo-api-...     persistent before requests\n```\n\nKnow for each job: what and why, location, state, expected result, dependents, revisit point, and code\u002Finput provenance. The helper persists operational metadata; the purpose and dependency graph remain in this agent's unified mental model.\n\nNatural scheduling points include after an edit, before a dependent decision, after another job completes, when no independent work remains, before final response, or when input may be needed. Do not repeatedly check a job without useful work between checks. If nothing useful remains, bounded waiting or inspection is honest and appropriate.\n\n## Use The Interface\n\nThe bundled script prefers isolated tmux sessions and falls back to background processes when tmux is unavailable. Run it by absolute path when necessary:\n\n```bash\ntimesharing\u002Fscripts\u002Ftimeshare.sh start tests \\\n  --purpose 'validate parser fix' --dependency 'review after serializer edit' \\\n  --provenance 'parser change A' -- npm test\ntimesharing\u002Fscripts\u002Ftimeshare.sh start api --persistent \\\n  --purpose 'serve integration target' -- npm run dev\ntimesharing\u002Fscripts\u002Ftimeshare.sh list\ntimesharing\u002Fscripts\u002Ftimeshare.sh status tests\ntimesharing\u002Fscripts\u002Ftimeshare.sh inspect tests --lines 80\ntimesharing\u002Fscripts\u002Ftimeshare.sh send repl --literal 'value.inspect' --enter\ntimesharing\u002Fscripts\u002Ftimeshare.sh interrupt api\ntimesharing\u002Fscripts\u002Ftimeshare.sh stop api\ntimesharing\u002Fscripts\u002Ftimeshare.sh cleanup\n```\n\nNames become lowercase `ts-*` job IDs. The tmux session also includes a repository name and workspace hash, preventing cross-repository collisions. Existing live jobs are never silently replaced. Resolve and inspect exact registered IDs before sending input, interrupting, or stopping anything.\n\nFor detailed operations, states, fallback behavior, and failures, read [references\u002Foperations.md](references\u002Foperations.md). For scheduling examples, read [references\u002Fexamples.md](references\u002Fexamples.md). Load the lower-level `tmux` skill only when pane layout or tmux-specific mechanics beyond this interface are needed.\n\n## Interpret State And Output\n\nThe lifecycle is:\n\n```text\nplanned -> running -> completed | failed | waiting | persistent\n        -> consumed -> cleaned up\n```\n\nThe helper detects running, completed, failed, persistent, and unexpectedly terminated. For interactive jobs, infer waiting versus computing from the recent tail and prompt. Infer a stall only from elapsed time, unchanged output, expected behavior, and process state; do not guess from silence alone.\n\nInspect recent output first. Increase `--lines` only when the error needs earlier context. Never flood this context with whole logs by default. Treat a missing completion marker as unknown\u002Funexpected termination, not success.\n\nBefore consuming a result, compare its launch provenance with current files\u002Fconfiguration. A run against state A does not validate later state B. Rerun relevant validation after material edits.\n\n## Interactive And Persistent Jobs\n\nMark servers\u002Fwatchers with `--persistent`; continued execution is expected, not failure. Mark REPLs\u002Fdebuggers with `--interactive`; tmux is required. Inspect the exact job before `send`, and send the least powerful input needed. Never send destructive keys to an uncertain target.\n\nThe shell fallback supports finite and persistent non-interactive commands, but not interactive input or durable terminal attachment. Report that limitation when fallback is used.\n\n## Finish\n\nBefore answering the user:\n\n- Inspect every important outstanding finite job.\n- Never claim success without a verified zero exit status.\n- Check that results still match the current code\u002Finput state.\n- State which requested services intentionally remain persistent.\n- Stop and clean disposable jobs you own; never kill uncertain or unrelated processes.\n- Preserve services the user explicitly asked to leave running.\n\nCommunicate only meaningful transitions: \"Tests are running while I inspect the serializer\" or \"The build passed; integration tests are still running.\" Do not narrate every tmux command or status check.\n\n## Anti-Patterns\n\n- **Subagent substitution:** same agent launches a process; never spawn another AI to run or interpret it.\n- **Fake concurrency:** do useful work after launch; do not launch and immediately poll in a loop.\n- **Context flooding:** read a relevant tail, not 50,000 repeated log lines.\n- **Excessive concurrency:** prefer 2-4 purposeful jobs, not 14 vague experiments.\n- **Stale validation:** record what state ran and never apply an old result to newer changes.\n- **Speculative overlap:** do not run work whose inputs are still being rewritten.\n","# Timesharing\n\nTimesharing lets one agent preserve one coherent reasoning context while opportunistically multiplexing its attention across independent external work.\n\nThe AI's reasoning stays serial. Only ordinary OS processes run concurrently. Never spawn a subagent, delegate interpretation, or create another AI context for this workflow.\n\n## Central Rule\n\n> Never block on an independent long-running process when useful independent work exists.\n\nTreat yourself as a cooperative scheduler:\n\n1. Identify a command whose result is not immediately required.\n2. Start it asynchronously and record its purpose, provenance, dependency, and revisit point.\n3. Continue concrete independent work immediately.\n4. Revisit at a natural boundary, not through busy polling.\n5. Interpret the result in the same context that launched it.\n\nUse a small number of jobs, usually 2-4. The goal is to hide external latency without losing cognitive coherence, not to maximize concurrency.\n\n## Decide\n\nTimeshare when judgment says all of these are true:\n\n- The command may take meaningful time or must remain alive.\n- Its result is not required for the next reasoning step.\n- Useful independent work exists, or later interaction\u002Fincremental output matters.\n- Its inputs are stable enough that the run will remain useful.\n\nTypical jobs include test suites, builds, installs, Docker builds, servers, watchers, long scripts, migrations, data\u002Fmodel processing, downloads, benchmarks, integration tests, REPLs, and debuggers.\n\nRun trivial or immediately blocking commands normally: `pwd`, `git status`, small reads, quick searches, and commands whose answer determines the very next step. Use judgment, not a hard timing threshold.\n\n## Schedule\n\nBefore launch, form a tiny dependency graph. Only overlap genuinely independent work. Do not start a run whose inputs are likely to be invalidated immediately.\n\nMaintain a lightweight in-context ledger:\n\n```text\nJOB       PURPOSE             LOCATION                 STATE    DEPENDS\u002FREVISIT\nts-tests  validate parser     tmux:ts-repo-tests-...   running  after serializer edit\nts-api    integration target  tmux:ts-repo-api-...     persistent before requests\n```\n\nKnow for each job: what and why, location, state, expected result, dependents, revisit point, and code\u002Finput provenance. The helper persists operational metadata; the purpose and dependency graph remain in this agent's unified mental model.\n\nNatural scheduling points include after an edit, before a dependent decision, after another job completes, when no independent work remains, before final response, or when input may be needed. Do not repeatedly check a job without useful work between checks. If nothing useful remains, bounded waiting or inspection is honest and appropriate.\n\n## Use The Interface\n\nThe bundled script prefers isolated tmux sessions and falls back to background processes when tmux is unavailable. Run it by absolute path when necessary:\n\n```bash\ntimesharing\u002Fscripts\u002Ftimeshare.sh start tests \\\n  --purpose 'validate parser fix' --dependency 'review after serializer edit' \\\n  --provenance 'parser change A' -- npm test\ntimesharing\u002Fscripts\u002Ftimeshare.sh start api --persistent \\\n  --purpose 'serve integration target' -- npm run dev\ntimesharing\u002Fscripts\u002Ftimeshare.sh list\ntimesharing\u002Fscripts\u002Ftimeshare.sh status tests\ntimesharing\u002Fscripts\u002Ftimeshare.sh inspect tests --lines 80\ntimesharing\u002Fscripts\u002Ftimeshare.sh send repl --literal 'value.inspect' --enter\ntimesharing\u002Fscripts\u002Ftimeshare.sh interrupt api\ntimesharing\u002Fscripts\u002Ftimeshare.sh stop api\ntimesharing\u002Fscripts\u002Ftimeshare.sh cleanup\n```\n\nNames become lowercase `ts-*` job IDs. The tmux session also includes a repository name and workspace hash, preventing cross-repository collisions. Existing live jobs are never silently replaced. Resolve and inspect exact registered IDs before sending input, interrupting, or stopping anything.\n\nFor detailed operations, states, fallback behavior, and failures, read [references\u002Foperations.md](references\u002Foperations.md). For scheduling examples, read [references\u002Fexamples.md](references\u002Fexamples.md). Load the lower-level `tmux` skill only when pane layout or tmux-specific mechanics beyond this interface are needed.\n\n## Interpret State And Output\n\nThe lifecycle is:\n\n```text\nplanned -> running -> completed | failed | waiting | persistent\n        -> consumed -> cleaned up\n```\n\nThe helper detects running, completed, failed, persistent, and unexpectedly terminated. For interactive jobs, infer waiting versus computing from the recent tail and prompt. Infer a stall only from elapsed time, unchanged output, expected behavior, and process state; do not guess from silence alone.\n\nInspect recent output first. Increase `--lines` only when the error needs earlier context. Never flood this context with whole logs by default. Treat a missing completion marker as unknown\u002Funexpected termination, not success.\n\nBefore consuming a result, compare its launch provenance with current files\u002Fconfiguration. A run against state A does not validate later state B. Rerun relevant validation after material edits.\n\n## Interactive And Persistent Jobs\n\nMark servers\u002Fwatchers with `--persistent`; continued execution is expected, not failure. Mark REPLs\u002Fdebuggers with `--interactive`; tmux is required. Inspect the exact job before `send`, and send the least powerful input needed. Never send destructive keys to an uncertain target.\n\nThe shell fallback supports finite and persistent non-interactive commands, but not interactive input or durable terminal attachment. Report that limitation when fallback is used.\n\n## Finish\n\nBefore answering the user:\n\n- Inspect every important outstanding finite job.\n- Never claim success without a verified zero exit status.\n- Check that results still match the current code\u002Finput state.\n- State which requested services intentionally remain persistent.\n- Stop and clean disposable jobs you own; never kill uncertain or unrelated processes.\n- Preserve services the user explicitly asked to leave running.\n\nCommunicate only meaningful transitions: \"Tests are running while I inspect the serializer\" or \"The build passed; integration tests are still running.\" Do not narrate every tmux command or status check.\n\n## Anti-Patterns\n\n- **Subagent substitution:** same agent launches a process; never spawn another AI to run or interpret it.\n- **Fake concurrency:** do useful work after launch; do not launch and immediately poll in a loop.\n- **Context flooding:** read a relevant tail, not 50,000 repeated log lines.\n- **Excessive concurrency:** prefer 2-4 purposeful jobs, not 14 vague experiments.\n- **Stale validation:** record what state ran and never apply an old result to newer changes.\n- **Speculative overlap:** do not run work whose inputs are still being rewritten.\n",{"name":5,"description":7,"compatibility":24},"Prefers tmux; supports a non-interactive shell-background fallback on macOS and Linux.","main","171fbdd3085ba76dacd148aa292d0559e8d913eb",1,"https:\u002F\u002Fgithub.com\u002Froerohan\u002Fskills","https:\u002F\u002Fgithub.com\u002Froerohan","https:\u002F\u002Froerohan.com","Engineer at @cloudflare | Founding Engineer at @dyte-in | CTFs | Ex-Tech Head at @csivitu (2020-21).","2026-09-04 18:05:21",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add roerohan\u002Fskills --skill timesharing",{"tool":39,"command":40},"bunx","bunx skills add roerohan\u002Fskills --skill timesharing",{"tool":42,"command":43},"pnpm","pnpm dlx skills add roerohan\u002Fskills --skill timesharing","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Froerohan\u002Fskills\u002Fblob\u002Fmain\u002Ftimesharing\u002FSKILL.md",true,[47,53,57],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fexamples.md","REFERENCE","参考文档",2989,"https:\u002F\u002Fgithub.com\u002Froerohan\u002Fskills\u002Fblob\u002Fmain\u002Ftimesharing\u002Freferences\u002Fexamples.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Foperations.md",3997,"https:\u002F\u002Fgithub.com\u002Froerohan\u002Fskills\u002Fblob\u002Fmain\u002Ftimesharing\u002Freferences\u002Foperations.md",{"path":58,"resourceType":59,"resourceTypeDesc":60,"sizeBytes":61,"sourceHtmlUrl":62},"scripts\u002Ftimeshare.sh","SCRIPT","脚本",14224,"https:\u002F\u002Fgithub.com\u002Froerohan\u002Fskills\u002Fblob\u002Fmain\u002Ftimesharing\u002Fscripts\u002Ftimeshare.sh",[64,71,78,85,92,99],{"skillNo":65,"name":66,"nameZh":6,"description":67,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":19},"01M72V6EQGTE5K3AP5B5AWJARB","why","Use for 'why does X work this way', 'why we picked Y', design rationale, regressions, postmortems, or data-backed thresholds. Discovers available MCPs and queries each evidence category (source control, issue tracker, long-form docs, real-time chat, infrastructure observability, error tracking, product analytics warehouse) in parallel, then returns a cited read on decisions and tradeoffs. Use how for runtime behavior.","https:\u002F\u002Fgithub.com\u002Froerohan\u002Fskills\u002Fblob\u002Fmain\u002Fwhy\u002FSKILL.md",[],[],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":19},"01M72V6B7459ZDJMS003D65S02","using-git-worktrees","Use when starting feature work that needs isolation from current workspace or before executing implementation plans - creates isolated git worktrees with smart directory selection and safety verification","https:\u002F\u002Fgithub.com\u002Froerohan\u002Fskills\u002Fblob\u002Fmain\u002Fusing-git-worktrees\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":19},"01M72V69RJ26MA6Q5RZ07XTG9V","unslop","Cut AI tells from any writing. Must always apply.","https:\u002F\u002Fgithub.com\u002Froerohan\u002Fskills\u002Fblob\u002Fmain\u002Funslop\u002FSKILL.md",[],[],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":19},"01M72V66RS1KZVXMB9NGC4RDH1","tmux","Manage long-running commands in the user's current tmux session. Use when starting dev servers, watchers, background logs, or reading\u002Finteracting with named tmux panes.","https:\u002F\u002Fgithub.com\u002Froerohan\u002Fskills\u002Fblob\u002Fmain\u002Ftmux\u002FSKILL.md",[],[],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":19},"01M72V62CFDFD6SJDA331JEWQ9","teach","Teach the user a new skill or concept, within this workspace.","https:\u002F\u002Fgithub.com\u002Froerohan\u002Fskills\u002Fblob\u002Fmain\u002Fteach\u002FSKILL.md",[],[],{"skillNo":100,"name":101,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":102,"categories":103,"keywords":104,"publishedAt":19},"01M72V5ZRWQWE16ZWWB1VXZ0NJ","spec-planner","https:\u002F\u002Fgithub.com\u002Froerohan\u002Fskills\u002Fblob\u002Fmain\u002Fspec-planner\u002FSKILL.md",[],[]]