[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EXM7DS2X7EC101D297DP5A":3,"skill-same:01M7EXM7DS2X7EC101D297DP5A":49},{"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":29,"commitSha":30,"skillDirectory":31,"forksCount":14,"repositoryHtmlUrl":32,"repositoryHomepageUrl":6,"ownerHtmlUrl":33,"ownerBlogUrl":6,"ownerDescription":34,"updatedAt":35,"install":36,"resources":48},"01M7EXM7DS2X7EC101D297DP5A","orch-plan-ralph",null,"Plexar orchestration skill \u002Forch-plan-ralph: \u002Forch-plan, then stand the whole thing up in one push","NovemberFalls\u002Fplexar-harness-oss","NovemberFalls","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F70782979?v=4","01M7EXEMTMCAQSKSJ6878BEERP","Leonard D Boord III","01M7EXEMV9Q5YT78A4HW17N891",0,"MIT","https:\u002F\u002Fgithub.com\u002FNovemberFalls\u002Fplexar-harness-oss\u002Fblob\u002Fmain\u002Fpackages\u002Fbundle\u002Fplexar\u002Fskills\u002Forch-plan-ralph\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: orch-plan-ralph\ndescription: \"Plexar orchestration skill \u002Forch-plan-ralph: \u002Forch-plan, then stand the whole thing up in one push\"\nuser-invocable: true\nmetadata:\n  source: \"~\u002F.claude\u002Fcommands\u002Forch-plan-ralph.md\"\n  source_sha256: \"6a261a21bf9336d81fb2fe3cf7e1a3f53dae1c792febbbda06cfafd440ff41b3\"\n  ported: \"2026-09-22\"\n---\n\n> **Running in Plexar Harness.** This skill was written for Claude Code; read its tool names through this map.\n> - `Agent` tool → the harness `subagent` tool (in-process child agent; `wait_agent` \u002F `list_agents` to collect results).\n> - Lane aliases `haiku` \u002F `sonnet` \u002F `opus` → there is one primary rig model (`Qwen\u002FQwen3.6-35B-A3B`; escalation `prism-ml\u002FTernary-Bonsai-2-27B`); express a lane as effort instead: MUNDANE → `off`, WORKHORSE and CRITICAL → `high`.\n> - `Bash` \u002F `PowerShell` → `bash` \u002F `pwsh`; `Read` \u002F `Write` \u002F `Edit` → `read` \u002F `write` \u002F `edit`.\n> - `plexar_agents` (ledger, cards) is Python in `\u003Cworkspace>\\plexar-framework`; call it through `pwsh`\u002F`bash` with that folder on `PYTHONPATH`.\n> - Anything this skill names that the harness lacks (e.g. `ToolSearch`, `Monitor`) is unavailable here; do that step by hand or skip it and say so.\n\n# \u002Forch-plan-ralph — \u002Forch-plan, then stand the whole thing up in one push\n\n**This is `\u002Forch-plan` with the ralph flag set. It is a shim, not a second planning skill.**\n\n> **Run `\u002Forch-plan` in full — all of it, unabbreviated.** Interrogation with cross-framing,\n> the approval questions, the nine-section `PLAN.md`, the decision log, milestones, QA\n> obligations, every invariant. Nothing about planning is shortened because a loop is\n> coming. A loop executing an un-interrogated backlog reaches the wrong destination faster\n> and with less supervision.\n\nThen do what is on this page, and only what is on this page. Two files that are \"the same\nexcept\" diverge — planning guidance lives in `\u002Forch-plan` and is never restated here.\n\n**Use it for pass 1 of a new or mostly-empty repo.** Standing up a skeleton is where\nper-node invocation costs the most and buys the least: the nodes are numerous, mostly\nindependent, mostly WORKHORSE\u002FMUNDANE. Hand-driving forty of those is ceremony.\n\n**Do not use it on a mature repo.** Where a change lands inside existing behavior and every\nnode carries real blast radius, use plain `\u002Forch-plan` and drive the nodes deliberately.\nA loop is a force multiplier in both directions.\n\n---\n\n## §1 · What the ralph flag adds to the plan\n\nTwo requirements the plan must satisfy before a loop may drive it:\n\n- **Every node needs a genuinely deterministic `accept` command.** Hand-driven, a human\n  notices a node that \"passed\" on a technicality. In a loop nobody is watching, so\n  **accept *is* the supervision**. A node whose accept is `echo done` will be reported\n  complete, and it will be wrong.\n- **The backlog must be dependency-sorted, unblocking-first.** A loop that hits a\n  dependency wall burns iterations discovering it.\n\n## §2 · Ask which mode — explain, do not choose\n\nPresent both and ask. This is the human's call:\n\n> **Attended** — each pass does a slice of the backlog, runs the gate, and stops, handing\n> you the diff and the gate result before anything else happens. You approve, then run the\n> next pass. You see every step before it compounds; a wrong turn costs one pass.\n>\n> **Unattended** — the loop runs itself until the backlog is done, a stop condition trips,\n> or a ceiling is hit. You start it and walk away. Far faster and far cheaper in your\n> attention; the cost is that a wrong turn in iteration 3 can be built on top of through\n> iteration 30 before you see it.\n>\n> **Guidance: attended for the first two or three passes of any repo**, until you have seen\n> what the loop actually does with your backlog — **then unattended for the long mechanical\n> middle**, where the nodes are numerous and boring and the gate is real. Unattended from a\n> cold start is how you get forty nodes of confidently wrong work.\n\nBoth modes emit the same harness; a flag selects behavior, and the human can switch at any\ntime. That is why both ship.\n\n## §3 · The file set — exact, so \"produces a ralph loop\" is verifiable\n\n```\nralph\u002F\n  PROMPT.md          invariant prompt, re-fed verbatim every iteration\n  loop.ps1           Windows driver\n  loop.sh            POSIX driver\n  STATE.md           done \u002F in progress \u002F blocked   (agent-owned; the loop's memory)\n  STOP               absent normally; create it to halt cleanly after the current pass\n  logs\u002Fiter-NNN.md   per pass: nodes attempted, gate result, diff stat, cost\n```\n\n### §3a · The two drivers share one serialization contract\n\n**Measured twice, 2026-08-06, and it cost a run each time.** `loop.ps1` and `loop.sh` read\nand write the same files, and PowerShell's defaults produce files bash cannot read:\n\n- `loop.ps1` written without a BOM is read as ANSI by PowerShell 5.1 — **seven parse\n  errors, the Windows driver does not run at all**, and in one cell it had already\n  corrupted `STATE.md` into mojibake.\n- `loop.ps1` wrote `ralph\u002Fcost.json` **with** a UTF-8 BOM; `loop.sh` could not parse it,\n  the budget flag came through empty, the CLI rejected it, and **two passes died in one\n  second each and tripped the no-progress halt.** The loop halted cleanly and wrote a\n  reason — it just wrote the *wrong* reason, which is worse than halting loudly.\n\n**Rules, because \"be careful with encodings\" is not a rule:**\n\n- Every file both drivers touch — `STATE.md`, `cost.txt`\u002F`cost.json`, the iteration logs —\n  is **UTF-8 without a BOM**. In PowerShell that means `[IO.File]::WriteAllText` with\n  `UTF8Encoding($false)`, not `Out-File` or `Set-Content`.\n- **`loop.ps1` itself carries a BOM**, or is ASCII-only. That is the opposite of the rule\n  above and both are needed: 5.1 needs the BOM to read the *script*; bash needs its absence\n  to read the *data*.\n- The harness's self-test must **round-trip one file through both drivers** before the first\n  real pass. Both defects above would have been caught by one such check.\n\n**`PROMPT.md` is invariant** — the same text every iteration. That is the whole technique.\nIt does not accumulate history, because the state that matters lives in `STATE.md` and in\nthe repo. A prompt that grows each pass is a different technique and it degrades.\n\nIts content, roughly:\n\n> Read `PLAN.md` and `ralph\u002FSTATE.md`. Pick the highest-priority unblocked, unfinished\n> nodes. Run `\u002Forch-code-anth` on them. Run each node's accept command. Update `STATE.md`\n> with what landed, what failed and why, and what is now unblocked. Commit each node as its\n> own candidate commit. If a milestone completed, write the alert into `STATE.md`. If you\n> are blocked on something only a human can answer, write it under BLOCKED and stop.\n\n## §4 · Stop conditions — all seven, whichever fires first\n\n1. **Done** — no unblocked, unfinished nodes remain.\n2. **`ralph\u002FSTOP` exists** — the human's clean brake. Checked *between* passes, so nothing\n   is interrupted mid-edit.\n3. **Max iterations** — hard cap, default 40.\n4. **Cost ceiling** — set by the human up front. Checked every iteration, non-negotiable.\n5. **No-progress** — two consecutive iterations with no node completed and no new commit.\n   The important one: **a stuck loop looks exactly like a working loop from outside**, and\n   without this check it burns the entire ceiling proving it.\n6. **Gate red at iteration start** — the tree was left broken. Stop; never build on it.\n7. **BLOCKED in `STATE.md`** — something only a human can answer.\n\nEvery halt writes its reason to `STATE.md` and the final log. **A loop that stops without\nsaying why is indistinguishable from one that finished.**\n\n## §5 · What the loop must never do\n\n- **Never touch a §6 destructive surface unattended.** Auth, money, concurrency,\n  migrations, production data: present the node as a plan with rollback and wait for a\n  human. **Skip it and continue with the rest of the backlog** rather than halting the whole\n  stand-up on one node. An autonomous loop is where the interlock matters most and where it\n  is most tempting to weaken it.\n- **Never push, merge, deploy, or release.** Local candidate commits only.\n- **Never mark a QA item passed.** Milestone completion writes the alert to `STATE.md`; QA\n  is its own phase with its own authority rules. A loop that grades its own work is the\n  failure mode this whole toolchain exists to prevent.\n- **Never rewrite a decision** in `PLAN.md` §3. It may append discovered nodes to the\n  backlog; changing a logged decision needs a human.\n\n## §6 · Reading the result\n\n1. `ralph\u002FSTATE.md` — what it thinks it did, and why it stopped.\n2. `git log --oneline` — what it actually did. **These two disagreeing is the finding.**\n3. The last two or three `ralph\u002Flogs\u002Fiter-*.md` — where it was when it stopped.\n4. Then `\u002Forch-qa` for completed milestones, and `\u002Forch-clean` if the run did renames,\n   signature migrations, or removals.\n\n**Expect to re-plan after pass 1.** A stand-up run's real output is a working skeleton *and\na much better-informed backlog* — the nodes that were harder than they looked, the\nassumptions that broke, the infrastructure marked UNKNOWN that is now known. `\u002Forch-plan`\nmerges on re-run rather than clobbering, so feeding that back is cheap and the second pass\nis far sharper than the first could have been.\n\n## Invariants\n\n- All `\u002Forch-plan` invariants hold, unabbreviated. This file adds; it never replaces.\n- No planning guidance is restated here. One engine, one flag.\n- Every node has a real deterministic accept command. In a loop, accept *is* supervision.\n- Both modes are offered and the difference is explained; the human chooses.\n- `PROMPT.md` is invariant across iterations. State lives in `STATE.md` and the repo.\n- All seven stop conditions are implemented, including no-progress and cost ceiling.\n  Every halt records its reason.\n- Destructive surfaces, publication, and QA promotion are never done by the loop, in any\n  mode.\n","> **Running in Plexar Harness.** This skill was written for Claude Code; read its tool names through this map.\n> - `Agent` tool → the harness `subagent` tool (in-process child agent; `wait_agent` \u002F `list_agents` to collect results).\n> - Lane aliases `haiku` \u002F `sonnet` \u002F `opus` → there is one primary rig model (`Qwen\u002FQwen3.6-35B-A3B`; escalation `prism-ml\u002FTernary-Bonsai-2-27B`); express a lane as effort instead: MUNDANE → `off`, WORKHORSE and CRITICAL → `high`.\n> - `Bash` \u002F `PowerShell` → `bash` \u002F `pwsh`; `Read` \u002F `Write` \u002F `Edit` → `read` \u002F `write` \u002F `edit`.\n> - `plexar_agents` (ledger, cards) is Python in `\u003Cworkspace>\\plexar-framework`; call it through `pwsh`\u002F`bash` with that folder on `PYTHONPATH`.\n> - Anything this skill names that the harness lacks (e.g. `ToolSearch`, `Monitor`) is unavailable here; do that step by hand or skip it and say so.\n\n# \u002Forch-plan-ralph — \u002Forch-plan, then stand the whole thing up in one push\n\n**This is `\u002Forch-plan` with the ralph flag set. It is a shim, not a second planning skill.**\n\n> **Run `\u002Forch-plan` in full — all of it, unabbreviated.** Interrogation with cross-framing,\n> the approval questions, the nine-section `PLAN.md`, the decision log, milestones, QA\n> obligations, every invariant. Nothing about planning is shortened because a loop is\n> coming. A loop executing an un-interrogated backlog reaches the wrong destination faster\n> and with less supervision.\n\nThen do what is on this page, and only what is on this page. Two files that are \"the same\nexcept\" diverge — planning guidance lives in `\u002Forch-plan` and is never restated here.\n\n**Use it for pass 1 of a new or mostly-empty repo.** Standing up a skeleton is where\nper-node invocation costs the most and buys the least: the nodes are numerous, mostly\nindependent, mostly WORKHORSE\u002FMUNDANE. Hand-driving forty of those is ceremony.\n\n**Do not use it on a mature repo.** Where a change lands inside existing behavior and every\nnode carries real blast radius, use plain `\u002Forch-plan` and drive the nodes deliberately.\nA loop is a force multiplier in both directions.\n\n---\n\n## §1 · What the ralph flag adds to the plan\n\nTwo requirements the plan must satisfy before a loop may drive it:\n\n- **Every node needs a genuinely deterministic `accept` command.** Hand-driven, a human\n  notices a node that \"passed\" on a technicality. In a loop nobody is watching, so\n  **accept *is* the supervision**. A node whose accept is `echo done` will be reported\n  complete, and it will be wrong.\n- **The backlog must be dependency-sorted, unblocking-first.** A loop that hits a\n  dependency wall burns iterations discovering it.\n\n## §2 · Ask which mode — explain, do not choose\n\nPresent both and ask. This is the human's call:\n\n> **Attended** — each pass does a slice of the backlog, runs the gate, and stops, handing\n> you the diff and the gate result before anything else happens. You approve, then run the\n> next pass. You see every step before it compounds; a wrong turn costs one pass.\n>\n> **Unattended** — the loop runs itself until the backlog is done, a stop condition trips,\n> or a ceiling is hit. You start it and walk away. Far faster and far cheaper in your\n> attention; the cost is that a wrong turn in iteration 3 can be built on top of through\n> iteration 30 before you see it.\n>\n> **Guidance: attended for the first two or three passes of any repo**, until you have seen\n> what the loop actually does with your backlog — **then unattended for the long mechanical\n> middle**, where the nodes are numerous and boring and the gate is real. Unattended from a\n> cold start is how you get forty nodes of confidently wrong work.\n\nBoth modes emit the same harness; a flag selects behavior, and the human can switch at any\ntime. That is why both ship.\n\n## §3 · The file set — exact, so \"produces a ralph loop\" is verifiable\n\n```\nralph\u002F\n  PROMPT.md          invariant prompt, re-fed verbatim every iteration\n  loop.ps1           Windows driver\n  loop.sh            POSIX driver\n  STATE.md           done \u002F in progress \u002F blocked   (agent-owned; the loop's memory)\n  STOP               absent normally; create it to halt cleanly after the current pass\n  logs\u002Fiter-NNN.md   per pass: nodes attempted, gate result, diff stat, cost\n```\n\n### §3a · The two drivers share one serialization contract\n\n**Measured twice, 2026-08-06, and it cost a run each time.** `loop.ps1` and `loop.sh` read\nand write the same files, and PowerShell's defaults produce files bash cannot read:\n\n- `loop.ps1` written without a BOM is read as ANSI by PowerShell 5.1 — **seven parse\n  errors, the Windows driver does not run at all**, and in one cell it had already\n  corrupted `STATE.md` into mojibake.\n- `loop.ps1` wrote `ralph\u002Fcost.json` **with** a UTF-8 BOM; `loop.sh` could not parse it,\n  the budget flag came through empty, the CLI rejected it, and **two passes died in one\n  second each and tripped the no-progress halt.** The loop halted cleanly and wrote a\n  reason — it just wrote the *wrong* reason, which is worse than halting loudly.\n\n**Rules, because \"be careful with encodings\" is not a rule:**\n\n- Every file both drivers touch — `STATE.md`, `cost.txt`\u002F`cost.json`, the iteration logs —\n  is **UTF-8 without a BOM**. In PowerShell that means `[IO.File]::WriteAllText` with\n  `UTF8Encoding($false)`, not `Out-File` or `Set-Content`.\n- **`loop.ps1` itself carries a BOM**, or is ASCII-only. That is the opposite of the rule\n  above and both are needed: 5.1 needs the BOM to read the *script*; bash needs its absence\n  to read the *data*.\n- The harness's self-test must **round-trip one file through both drivers** before the first\n  real pass. Both defects above would have been caught by one such check.\n\n**`PROMPT.md` is invariant** — the same text every iteration. That is the whole technique.\nIt does not accumulate history, because the state that matters lives in `STATE.md` and in\nthe repo. A prompt that grows each pass is a different technique and it degrades.\n\nIts content, roughly:\n\n> Read `PLAN.md` and `ralph\u002FSTATE.md`. Pick the highest-priority unblocked, unfinished\n> nodes. Run `\u002Forch-code-anth` on them. Run each node's accept command. Update `STATE.md`\n> with what landed, what failed and why, and what is now unblocked. Commit each node as its\n> own candidate commit. If a milestone completed, write the alert into `STATE.md`. If you\n> are blocked on something only a human can answer, write it under BLOCKED and stop.\n\n## §4 · Stop conditions — all seven, whichever fires first\n\n1. **Done** — no unblocked, unfinished nodes remain.\n2. **`ralph\u002FSTOP` exists** — the human's clean brake. Checked *between* passes, so nothing\n   is interrupted mid-edit.\n3. **Max iterations** — hard cap, default 40.\n4. **Cost ceiling** — set by the human up front. Checked every iteration, non-negotiable.\n5. **No-progress** — two consecutive iterations with no node completed and no new commit.\n   The important one: **a stuck loop looks exactly like a working loop from outside**, and\n   without this check it burns the entire ceiling proving it.\n6. **Gate red at iteration start** — the tree was left broken. Stop; never build on it.\n7. **BLOCKED in `STATE.md`** — something only a human can answer.\n\nEvery halt writes its reason to `STATE.md` and the final log. **A loop that stops without\nsaying why is indistinguishable from one that finished.**\n\n## §5 · What the loop must never do\n\n- **Never touch a §6 destructive surface unattended.** Auth, money, concurrency,\n  migrations, production data: present the node as a plan with rollback and wait for a\n  human. **Skip it and continue with the rest of the backlog** rather than halting the whole\n  stand-up on one node. An autonomous loop is where the interlock matters most and where it\n  is most tempting to weaken it.\n- **Never push, merge, deploy, or release.** Local candidate commits only.\n- **Never mark a QA item passed.** Milestone completion writes the alert to `STATE.md`; QA\n  is its own phase with its own authority rules. A loop that grades its own work is the\n  failure mode this whole toolchain exists to prevent.\n- **Never rewrite a decision** in `PLAN.md` §3. It may append discovered nodes to the\n  backlog; changing a logged decision needs a human.\n\n## §6 · Reading the result\n\n1. `ralph\u002FSTATE.md` — what it thinks it did, and why it stopped.\n2. `git log --oneline` — what it actually did. **These two disagreeing is the finding.**\n3. The last two or three `ralph\u002Flogs\u002Fiter-*.md` — where it was when it stopped.\n4. Then `\u002Forch-qa` for completed milestones, and `\u002Forch-clean` if the run did renames,\n   signature migrations, or removals.\n\n**Expect to re-plan after pass 1.** A stand-up run's real output is a working skeleton *and\na much better-informed backlog* — the nodes that were harder than they looked, the\nassumptions that broke, the infrastructure marked UNKNOWN that is now known. `\u002Forch-plan`\nmerges on re-run rather than clobbering, so feeding that back is cheap and the second pass\nis far sharper than the first could have been.\n\n## Invariants\n\n- All `\u002Forch-plan` invariants hold, unabbreviated. This file adds; it never replaces.\n- No planning guidance is restated here. One engine, one flag.\n- Every node has a real deterministic accept command. In a loop, accept *is* supervision.\n- Both modes are offered and the difference is explained; the human chooses.\n- `PROMPT.md` is invariant across iterations. State lives in `STATE.md` and the repo.\n- All seven stop conditions are implemented, including no-progress and cost ceiling.\n  Every halt records its reason.\n- Destructive surfaces, publication, and QA promotion are never done by the loop, in any\n  mode.\n",{"name":5,"metadata":24,"description":7,"user-invocable":28},{"ported":25,"source":26,"source_sha256":27},"2026-09-22","~\u002F.claude\u002Fcommands\u002Forch-plan-ralph.md","6a261a21bf9336d81fb2fe3cf7e1a3f53dae1c792febbbda06cfafd440ff41b3",true,"main","90e8ac241399a5c822403effcc2a15eb13dd2670","packages\u002Fbundle\u002Fplexar\u002Fskills\u002Forch-plan-ralph","https:\u002F\u002Fgithub.com\u002FNovemberFalls\u002Fplexar-harness-oss","https:\u002F\u002Fgithub.com\u002FNovemberFalls","Aspiring Programmer.","2026-10-22 04:07:59",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":28},[38,41,44],{"tool":39,"command":40},"npx","npx skills add NovemberFalls\u002Fplexar-harness-oss --skill orch-plan-ralph",{"tool":42,"command":43},"bunx","bunx skills add NovemberFalls\u002Fplexar-harness-oss --skill orch-plan-ralph",{"tool":45,"command":46},"pnpm","pnpm dlx skills add NovemberFalls\u002Fplexar-harness-oss --skill orch-plan-ralph","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FNovemberFalls\u002Fplexar-harness-oss\u002Fblob\u002Fmain\u002Fpackages\u002Fbundle\u002Fplexar\u002Fskills\u002Forch-plan-ralph\u002FSKILL.md",[],[50,57,64,71,78,85],{"skillNo":51,"name":52,"nameZh":6,"description":53,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":54,"categories":55,"keywords":56,"publishedAt":19},"01M7EXNGMA687XV7VCQ1K5QWRR","user-only-skill","Prove model-disabled skills stay outside the model catalog.","https:\u002F\u002Fgithub.com\u002FNovemberFalls\u002Fplexar-harness-oss\u002Fblob\u002Fmain\u002Fsnapshots\u002Fsession\u002Fskill-load\u002Fworkspace\u002F.dsh\u002Fskills\u002Fuser-only-skill\u002FSKILL.md",[],[],{"skillNo":58,"name":59,"nameZh":6,"description":60,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":61,"categories":62,"keywords":63,"publishedAt":19},"01M7EXNF0426EZM03WC5GNQ0EF","snapshot-skill","Exercise project skill discovery and loading in snapshot tests.","https:\u002F\u002Fgithub.com\u002FNovemberFalls\u002Fplexar-harness-oss\u002Fblob\u002Fmain\u002Fsnapshots\u002Fsession\u002Fskill-load\u002Fworkspace\u002F.dsh\u002Fskills\u002Fsnapshot-skill\u002FSKILL.md",[],[],{"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},"01M7EXNDR0S4WPJ63KYX40BFZE","model-only-skill","Prove user-disabled skills remain available to the model.","https:\u002F\u002Fgithub.com\u002FNovemberFalls\u002Fplexar-harness-oss\u002Fblob\u002Fmain\u002Fsnapshots\u002Fsession\u002Fskill-load\u002Fworkspace\u002F.dsh\u002Fskills\u002Fmodel-only-skill\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},"01M7EXNBP1V9VKG6ZTN69FGHBB","office-xlsx","Read, create, and modify Excel workbooks (.xlsx), including cell values, formulas, formatting, and pandas analysis. Use when an Excel workbook is an input or deliverable.","https:\u002F\u002Fgithub.com\u002FNovemberFalls\u002Fplexar-harness-oss\u002Fblob\u002Fmain\u002Fpackages\u002Fskill\u002Fskill-office\u002Fassets\u002Foffice-xlsx\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},"01M7EXN9D1AKV8E92E4P1GFJVH","office-pptx","Create, read, edit, and check PowerPoint presentations (.pptx), including slide text, tables, images, and charts. Use when a PPTX file is an input or requested deliverable.","https:\u002F\u002Fgithub.com\u002FNovemberFalls\u002Fplexar-harness-oss\u002Fblob\u002Fmain\u002Fpackages\u002Fskill\u002Fskill-office\u002Fassets\u002Foffice-pptx\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},"01M7EXN78JHZQB7PMVQ3TPQ53B","office-docx","Create, read, edit, and check Word documents (.docx), including reports, letters, and formatted tables. Use when a DOCX file is an input or requested deliverable.","https:\u002F\u002Fgithub.com\u002FNovemberFalls\u002Fplexar-harness-oss\u002Fblob\u002Fmain\u002Fpackages\u002Fskill\u002Fskill-office\u002Fassets\u002Foffice-docx\u002FSKILL.md",[],[]]