[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6DAYHCYCVHZ4QN5ZWZ5WRBB":3,"skill-same:01M6DAYHCYCVHZ4QN5ZWZ5WRBB":91},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":22,"publishedAt":43,"linkOnly":44,"markdownContent":45,"markdownBody":46,"frontMatter":47,"refName":55,"commitSha":56,"skillDirectory":57,"forksCount":13,"repositoryHtmlUrl":58,"repositoryHomepageUrl":6,"ownerHtmlUrl":59,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":60,"install":61,"resources":74},"01M6DAYHCYCVHZ4QN5ZWZ5WRBB","user-run-scripts",null,"Use whenever a task needs something the agent's own sandbox cannot do — an interactive sudo\u002Froot password, another TTY prompt, an action blocked by sandbox or permission policy, or any manual out-of-band step (e.g. a 2FA code) — by handing the user a script to run themselves and reading its output back from disk instead of a copy\u002Fpaste round-trip. Not for commands the agent can already run itself. Triggers: \"I need sudo for this\", \"this needs your password\", \"run this yourself\", \"needs an interactive terminal\", \"sandbox won't let me do X\".","tzeusy-org\u002Fai-bootstrap","tzeusy-org","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F141543652?v=4","01M6DAM1AP09H27BYGFY9QCNPB","01M6DAM1F4R187EK11XDN9JZKH",0,"MIT","https:\u002F\u002Fgithub.com\u002Ftzeusy-org\u002Fai-bootstrap\u002Fblob\u002Fmain\u002Fskills\u002Fpersonal\u002Fth-tooling\u002Fsubskills\u002Fuser-run-scripts\u002FSKILL.md",[17],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42],"user","user run scripts","run","use","sandbox","triggers","tee","active","me","back","task","do","policy","linux","interactive","prompt","password","Bash","disk","code","2026-11-10 10:12:30",false,"---\nname: user-run-scripts\ndescription: >\n  Use whenever a task needs something the agent's own sandbox cannot do —\n  an interactive sudo\u002Froot password, another TTY prompt, an action blocked\n  by sandbox or permission policy, or any manual out-of-band step (e.g. a\n  2FA code) — by handing the user a script to run themselves and reading\n  its output back from disk instead of a copy\u002Fpaste round-trip. Not for\n  commands the agent can already run itself. Triggers: \"I need sudo for\n  this\", \"this needs your password\", \"run this yourself\", \"needs an\n  interactive terminal\", \"sandbox won't let me do X\".\nmetadata:\n  owner: tze\n  authors:\n    - tze\n    - Claude Sonnet 5\n  status: active\n  last_reviewed: \"2026-08-30\"\ncompatibility: Linux\u002FmacOS, bash, coreutils (tee, ln -s, date); assumes an interactive sudo-capable user terminal separate from the agent's own shell.\n---\n\n# User-Run Scripts\n\nThe agent's shell often cannot do what a task needs: `sudo` prompts for a\npassword with no TTY, some actions are blocked by sandbox\u002Fpermission\npolicy, or a step is inherently manual (type a 2FA code, click a browser\nconsent screen). The fix is never to work around the credential — it is to\nwrite the user a script, have them run it in their own terminal, and read\nthe results back from disk.\n\n## Do not use this skill for\n\nRoutine commands the agent's own tools can already run. Reach for this only\nwhen a specific step is blocked — not as a default way to hand off work.\n\n## Never obtain or type the user's password\n\nNo `sudo -S` with a piped password, no askpass workarounds, no asking the\nuser to paste a password into chat. If a step needs credentials, the script\ngoes to the user's terminal — always.\n\n## The output convention (load-bearing part of this skill)\n\nEvery script writes into `~\u002F.tmp\u002F{slug}\u002F`, where `{slug}` is a short\nkebab-case name chosen for the task (e.g. `root-disk-audit`):\n\n- a fresh timestamped subdirectory per run — `~\u002F.tmp\u002F{slug}\u002F\u003CYYYYmmdd-HHMMSS>\u002F`\n  — with every command's stdout\u002Fstderr `tee`'d into a file under it\n- a `latest` symlink at `~\u002F.tmp\u002F{slug}\u002Flatest`, repointed with `ln -sfn` at\n  the end of a successful run, so history survives but the newest run is\n  always at a fixed, timestamp-free path\n\n```bash\nslug=\"CHANGE_ME\"\nreal_home=\"$(getent passwd \"${SUDO_USER:-$USER}\" | cut -d: -f6)\"  # see pitfall below\ntmproot=\"$real_home\u002F.tmp\u002F$slug\"\noutdir=\"$tmproot\u002F$(date +%Y%m%d-%H%M%S)\"\nmkdir -p \"$outdir\"\n\necho \"==> some diagnostic\" | tee \"$outdir\u002Fsome-diagnostic.txt\"\n# ... more commands, each teed into its own file ...\n\nln -sfn \"$outdir\" \"$tmproot\u002Flatest\"   # always last, on success\n[ \"$(id -u)\" -eq 0 ] && [ -n \"${SUDO_USER:-}\" ] && chown -R \"$SUDO_USER\" \"$tmproot\"\n```\n\nA ready-to-copy version of this scaffold is\n[`scripts\u002Ftemplate.sh`](.\u002Fscripts\u002Ftemplate.sh) — start from it rather than\nretyping the boilerplate each time.\n\nOnce the user confirms the run finished, resolve `~\u002F.tmp\u002F{slug}\u002Flatest\u002F`\nyourself (`ls`, `readlink -f`, or `Read` the files under it directly).\n**Never ask the user to paste output back into chat.**\n\n## Mandatory pre-presentation risk summary\n\nBefore showing the script or telling the user to run it, always disclose,\nin three labeled lines:\n\n```\nWhat it does: \u003Cplain-language summary>\nRisk: \u003Cdestructive ops, irreversible deletes, restarts, config changes — or \"none\">\nState changes: \u003Cfiles\u002FDBs\u002Fk8s resources touched, packages installed — or \"none, read-only\">\n```\n\nNever skip this step. A purely read-only script (`du`, `df`, `lsof`) still\ngets the disclosure — say \"none\" \u002F \"none, read-only\" explicitly rather than\nomitting the section.\n\n## Worked example\n\n[`references\u002Fworked-example-root-disk-audit.md`](.\u002Freferences\u002Fworked-example-root-disk-audit.md)\nwalks a real case: auditing root filesystem usage that required `sudo du`\ninto root-owned `\u002Fvar\u002Flib\u002Francher` (containerd + Kubernetes PV storage),\nwith the full script, the disclosure given to the user, and how the results\nwere read back from `~\u002F.tmp\u002Froot-disk-audit\u002Flatest\u002F` with no paste-back.\n\n## Pitfall: `sudo` resets `$HOME`\n\nHandoffs commonly say \"run `sudo ~\u002Fscript.sh`\" (whole-script sudo, simplest\nfor the user — one password prompt instead of one per line). But `sudo`\nresets `$HOME` to the target user's home (`\u002Froot`) by default, so a bare\n`$HOME\u002F.tmp\u002F...` inside the script silently writes into `\u002Froot\u002F.tmp\u002F`,\nunreadable by the agent — and by the user without another sudo. This bit a\nreal run: `sudo ~\u002Froot-disk-audit.sh` wrote its whole output under\n`\u002Froot\u002F.tmp\u002Froot-disk-audit-*\u002F` instead of `~\u002F.tmp\u002F`.\n\nAlways resolve the real invoking user's home explicitly, and hand ownership\nback if the script ran as root — both are in the template and the scaffold\nabove (`real_home=\"$(getent passwd \"${SUDO_USER:-$USER}\" | cut -d: -f6)\"`,\nthen `chown -R \"$SUDO_USER\" \"$tmproot\"` at the end when `id -u` is 0). Do\nthis even when individual commands also say `sudo` themselves — the whole\nscript may still be invoked under an outer `sudo`.\n\n## Hard stops\n\n- Never embed a password or secret in the generated script.\n- Never attempt the privileged operation directly yourself, even if some\n  sandbox permission would technically allow a piece of it — if the task\n  fundamentally needs the user's own credentials or an interactive prompt,\n  the whole thing goes in the script for them to run.\n- Default to `set -uo pipefail` (not `-e`) so a diagnostic script keeps\n  reporting later sections after one command fails. Use `set -euo pipefail`\n  only for a sequence of dependent mutating steps where any failure must\n  halt immediately — say so in the script when you do.\n","# User-Run Scripts\n\nThe agent's shell often cannot do what a task needs: `sudo` prompts for a\npassword with no TTY, some actions are blocked by sandbox\u002Fpermission\npolicy, or a step is inherently manual (type a 2FA code, click a browser\nconsent screen). The fix is never to work around the credential — it is to\nwrite the user a script, have them run it in their own terminal, and read\nthe results back from disk.\n\n## Do not use this skill for\n\nRoutine commands the agent's own tools can already run. Reach for this only\nwhen a specific step is blocked — not as a default way to hand off work.\n\n## Never obtain or type the user's password\n\nNo `sudo -S` with a piped password, no askpass workarounds, no asking the\nuser to paste a password into chat. If a step needs credentials, the script\ngoes to the user's terminal — always.\n\n## The output convention (load-bearing part of this skill)\n\nEvery script writes into `~\u002F.tmp\u002F{slug}\u002F`, where `{slug}` is a short\nkebab-case name chosen for the task (e.g. `root-disk-audit`):\n\n- a fresh timestamped subdirectory per run — `~\u002F.tmp\u002F{slug}\u002F\u003CYYYYmmdd-HHMMSS>\u002F`\n  — with every command's stdout\u002Fstderr `tee`'d into a file under it\n- a `latest` symlink at `~\u002F.tmp\u002F{slug}\u002Flatest`, repointed with `ln -sfn` at\n  the end of a successful run, so history survives but the newest run is\n  always at a fixed, timestamp-free path\n\n```bash\nslug=\"CHANGE_ME\"\nreal_home=\"$(getent passwd \"${SUDO_USER:-$USER}\" | cut -d: -f6)\"  # see pitfall below\ntmproot=\"$real_home\u002F.tmp\u002F$slug\"\noutdir=\"$tmproot\u002F$(date +%Y%m%d-%H%M%S)\"\nmkdir -p \"$outdir\"\n\necho \"==> some diagnostic\" | tee \"$outdir\u002Fsome-diagnostic.txt\"\n# ... more commands, each teed into its own file ...\n\nln -sfn \"$outdir\" \"$tmproot\u002Flatest\"   # always last, on success\n[ \"$(id -u)\" -eq 0 ] && [ -n \"${SUDO_USER:-}\" ] && chown -R \"$SUDO_USER\" \"$tmproot\"\n```\n\nA ready-to-copy version of this scaffold is\n[`scripts\u002Ftemplate.sh`](.\u002Fscripts\u002Ftemplate.sh) — start from it rather than\nretyping the boilerplate each time.\n\nOnce the user confirms the run finished, resolve `~\u002F.tmp\u002F{slug}\u002Flatest\u002F`\nyourself (`ls`, `readlink -f`, or `Read` the files under it directly).\n**Never ask the user to paste output back into chat.**\n\n## Mandatory pre-presentation risk summary\n\nBefore showing the script or telling the user to run it, always disclose,\nin three labeled lines:\n\n```\nWhat it does: \u003Cplain-language summary>\nRisk: \u003Cdestructive ops, irreversible deletes, restarts, config changes — or \"none\">\nState changes: \u003Cfiles\u002FDBs\u002Fk8s resources touched, packages installed — or \"none, read-only\">\n```\n\nNever skip this step. A purely read-only script (`du`, `df`, `lsof`) still\ngets the disclosure — say \"none\" \u002F \"none, read-only\" explicitly rather than\nomitting the section.\n\n## Worked example\n\n[`references\u002Fworked-example-root-disk-audit.md`](.\u002Freferences\u002Fworked-example-root-disk-audit.md)\nwalks a real case: auditing root filesystem usage that required `sudo du`\ninto root-owned `\u002Fvar\u002Flib\u002Francher` (containerd + Kubernetes PV storage),\nwith the full script, the disclosure given to the user, and how the results\nwere read back from `~\u002F.tmp\u002Froot-disk-audit\u002Flatest\u002F` with no paste-back.\n\n## Pitfall: `sudo` resets `$HOME`\n\nHandoffs commonly say \"run `sudo ~\u002Fscript.sh`\" (whole-script sudo, simplest\nfor the user — one password prompt instead of one per line). But `sudo`\nresets `$HOME` to the target user's home (`\u002Froot`) by default, so a bare\n`$HOME\u002F.tmp\u002F...` inside the script silently writes into `\u002Froot\u002F.tmp\u002F`,\nunreadable by the agent — and by the user without another sudo. This bit a\nreal run: `sudo ~\u002Froot-disk-audit.sh` wrote its whole output under\n`\u002Froot\u002F.tmp\u002Froot-disk-audit-*\u002F` instead of `~\u002F.tmp\u002F`.\n\nAlways resolve the real invoking user's home explicitly, and hand ownership\nback if the script ran as root — both are in the template and the scaffold\nabove (`real_home=\"$(getent passwd \"${SUDO_USER:-$USER}\" | cut -d: -f6)\"`,\nthen `chown -R \"$SUDO_USER\" \"$tmproot\"` at the end when `id -u` is 0). Do\nthis even when individual commands also say `sudo` themselves — the whole\nscript may still be invoked under an outer `sudo`.\n\n## Hard stops\n\n- Never embed a password or secret in the generated script.\n- Never attempt the privileged operation directly yourself, even if some\n  sandbox permission would technically allow a piece of it — if the task\n  fundamentally needs the user's own credentials or an interactive prompt,\n  the whole thing goes in the script for them to run.\n- Default to `set -uo pipefail` (not `-e`) so a diagnostic script keeps\n  reporting later sections after one command fails. Use `set -euo pipefail`\n  only for a sequence of dependent mutating steps where any failure must\n  halt immediately — say so in the script when you do.\n",{"name":5,"metadata":48,"description":53,"compatibility":54},{"owner":49,"status":30,"authors":50,"last_reviewed":52},"tze",[49,51],"Claude Sonnet 5","2026-08-30","Use whenever a task needs something the agent's own sandbox cannot do — an interactive sudo\u002Froot password, another TTY prompt, an action blocked by sandbox or permission policy, or any manual out-of-band step (e.g. a 2FA code) — by handing the user a script to run themselves and reading its output back from disk instead of a copy\u002Fpaste round-trip. Not for commands the agent can already run itself. Triggers: \"I need sudo for this\", \"this needs your password\", \"run this yourself\", \"needs an interactive terminal\", \"sandbox won't let me do X\".\n","Linux\u002FmacOS, bash, coreutils (tee, ln -s, date); assumes an interactive sudo-capable user terminal separate from the agent's own shell.","main","c907d6f0bf96bd61b936e8972e0f1c66ebb8148c","skills\u002Fpersonal\u002Fth-tooling\u002Fsubskills\u002Fuser-run-scripts","https:\u002F\u002Fgithub.com\u002Ftzeusy-org\u002Fai-bootstrap","https:\u002F\u002Fgithub.com\u002Ftzeusy-org","2026-11-10 23:17:25",{"source":8,"skillName":5,"commands":62,"prompt":72,"zipAvailable":73},[63,66,69],{"tool":64,"command":65},"npx","npx skills add tzeusy-org\u002Fai-bootstrap --skill user-run-scripts",{"tool":67,"command":68},"bunx","bunx skills add tzeusy-org\u002Fai-bootstrap --skill user-run-scripts",{"tool":70,"command":71},"pnpm","pnpm dlx skills add tzeusy-org\u002Fai-bootstrap --skill user-run-scripts","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Ftzeusy-org\u002Fai-bootstrap\u002Fblob\u002Fmain\u002Fskills\u002Fpersonal\u002Fth-tooling\u002Fsubskills\u002Fuser-run-scripts\u002FSKILL.md",true,[75,81,85],{"path":76,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":79,"sourceHtmlUrl":80},"references\u002Fworked-example-root-disk-audit.md","REFERENCE","参考文档",3001,"https:\u002F\u002Fgithub.com\u002Ftzeusy-org\u002Fai-bootstrap\u002Fblob\u002Fmain\u002Fskills\u002Fpersonal\u002Fth-tooling\u002Fsubskills\u002Fuser-run-scripts\u002Freferences\u002Fworked-example-root-disk-audit.md",{"path":82,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":83,"sourceHtmlUrl":84},"resources\u002Fimplementation-playbook.md",42170,"https:\u002F\u002Fgithub.com\u002Fagency-black\u002FSkills\u002Fblob\u002Fmain\u002FCopia de api-testing-observability-api-mock\u002Fresources\u002Fimplementation-playbook.md",{"path":86,"resourceType":87,"resourceTypeDesc":88,"sizeBytes":89,"sourceHtmlUrl":90},"scripts\u002Ftemplate.sh","SCRIPT","脚本",1726,"https:\u002F\u002Fgithub.com\u002Ftzeusy-org\u002Fai-bootstrap\u002Fblob\u002Fmain\u002Fskills\u002Fpersonal\u002Fth-tooling\u002Fsubskills\u002Fuser-run-scripts\u002Fscripts\u002Ftemplate.sh",[92,113,137,152,171,182],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":43},"01M6DAZD21TWH2ET70V9FF75CQ","user-run","Use when a step needs privileges this session lacks (docker, sudo, another user's checkout or secrets, host access) and the owner must run it as tze. Writes a reviewable \u002Fhome\u002Forca\u002F.tmp\u002F\u003Cslug>\u002Frun.sh that impersonates tze (USER_TO_IMPERSONATE overrides), tees timestamped logs plus latest.log and collects artifacts there; Claude presents it with a risk and time brief, then continues from the logs. Triggers: \"I'll run it for you\", \"permission denied on docker.sock\", \"needs sudo\", \"sudo: a password is required\", \"run as tze\", \"\u002Fuser-run\".","https:\u002F\u002Fgithub.com\u002Ftzeusy-org\u002Fai-bootstrap\u002Fblob\u002Fmain\u002Fskills\u002Fpersonal\u002Fuser-run\u002FSKILL.md",[],[99,23,25,100,101,26,102,103,28,104,30,105,106,107,108,39,109,110,111,112],"user run","docker","logs","host","as","permission","time","log","artifacts","session","access","secrets","risk","brief",{"skillNo":114,"name":115,"nameZh":6,"description":116,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":127,"publishedAt":43},"01M6DAZAHTCBAD2WF0GAHWPST8","requesting-code-review","Use when completing tasks, implementing major features, or before merging to verify work meets requirements","https:\u002F\u002Fgithub.com\u002Ftzeusy-org\u002Fai-bootstrap\u002Fblob\u002Fmain\u002Fskills\u002Fpersonal\u002Fth-workflow\u002Fsubskills\u002Frequesting-code-review\u002FSKILL.md",[119,122],{"categoryNo":120,"categoryName":121,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000007","代码审查",{"categoryNo":123,"categoryName":124,"taxonomyType":125,"taxonomyTypeDesc":126},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[128,42,129,130,131,132,30,133,134,26,135,136],"requesting code review","review","code review","work","requirements","upstream","superpowers","verify","tasks",{"skillNo":138,"name":139,"nameZh":6,"description":140,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":141,"categories":142,"keywords":145,"publishedAt":43},"01M6DAZ7CJ82D81XZB2YHVVY48","receiving-code-review","Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation","https:\u002F\u002Fgithub.com\u002Ftzeusy-org\u002Fai-bootstrap\u002Fblob\u002Fmain\u002Fskills\u002Fpersonal\u002Fth-workflow\u002Fsubskills\u002Freceiving-code-review\u002FSKILL.md",[143,144],{"categoryNo":120,"categoryName":121,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":123,"categoryName":124,"taxonomyType":125,"taxonomyTypeDesc":126},[146,42,129,130,30,133,147,134,26,148,149,150,151],"receiving code review","implementation","code review feedback","technical","feedback","verification",{"skillNo":153,"name":154,"nameZh":6,"description":155,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":156,"categories":157,"keywords":158,"publishedAt":43},"01M6DAZ4ER76ADXJ9KRFP7V3Q9","finishing-a-development-branch","Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup","https:\u002F\u002Fgithub.com\u002Ftzeusy-org\u002Fai-bootstrap\u002Fblob\u002Fmain\u002Fskills\u002Fpersonal\u002Fth-workflow\u002Fsubskills\u002Ffinishing-a-development-branch\u002FSKILL.md",[],[159,160,161,162,26,163,131,164,165,166,134,167,168,169,30,133,170,147],"branch","development","finishing a development branch","integrate","decide","how","cleanup","pr","merge","tests","structured","options",{"skillNo":172,"name":173,"nameZh":6,"description":174,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":175,"categories":176,"keywords":177,"publishedAt":43},"01M6DAYZFWWW2YPKP89QNQR96H","executing-plans","Use when you have a written implementation plan to execute in a separate session with review checkpoints","https:\u002F\u002Fgithub.com\u002Ftzeusy-org\u002Fai-bootstrap\u002Fblob\u002Fmain\u002Fskills\u002Fpersonal\u002Fth-workflow\u002Fsubskills\u002Fexecuting-plans\u002FSKILL.md",[],[178,179,180,129,181,30,133,147,134,26,108],"plans","executing plans","plan","execute",{"skillNo":183,"name":184,"nameZh":6,"description":185,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":186,"categories":187,"keywords":188,"publishedAt":43},"01M6DAYWM2YQHYCK7C35K1D29J","dispatching-parallel-agents","Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies","https:\u002F\u002Fgithub.com\u002Ftzeusy-org\u002Fai-bootstrap\u002Fblob\u002Fmain\u002Fskills\u002Fpersonal\u002Fth-workflow\u002Fsubskills\u002Fdispatching-parallel-agents\u002FSKILL.md",[],[189,190,191,26,134,133,192,30,193,136,194],"parallel agents","parallel","dispatching parallel agents","state","dependencies","shared"]