[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M77HGKV2EJS0NW7BS3P9BSW3":3,"skill-same:01M77HGKV2EJS0NW7BS3P9BSW3":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":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":32,"ownerHtmlUrl":33,"ownerBlogUrl":34,"ownerDescription":6,"updatedAt":35,"install":36,"resources":48},"01M77HGKV2EJS0NW7BS3P9BSW3","opencode-wave",null,"Spawn and coordinate deterministic, collision-safe OpenCode leaf worker waves with high reasoning effort (--variant high), automated approvals (--auto), mandatory detached worktree isolation (--worktree), pre-flight lane lease arbitration, and automated closed-loop harvest.","anthony-chaudhary\u002Ffak","anthony-chaudhary","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F18080164?v=4","01M77GSKCAEKV3JS07J40B94DY","Anthony Chaudhary","01M77GSKD26Z7F4FQQD4VA30VX",48,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fanthony-chaudhary\u002Ffak\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fopencode-wave\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: opencode-wave\ndescription: Spawn and coordinate deterministic, collision-safe OpenCode leaf worker waves with high reasoning effort (--variant high), automated approvals (--auto), mandatory detached worktree isolation (--worktree), pre-flight lane lease arbitration, and automated closed-loop harvest.\ndisable-model-invocation: false\nuser-invocable: true\nallowed-tools: Read, Edit, Write, Grep, Glob, Bash, Task\nargument-hint: \"[--top 10] [--wave-size 4] [--variant high] [--worktree] [--workspace \u003Cpath>] [--from-issues \u003Cfile>] [--harvest]\"\n---\n\n# \u002Fopencode-wave — High-Effort Deterministic OpenCode Multi-Agent Wave Dispatch\n\nThe campaign coordinator for running multi-issue resolution waves in OpenCode. It takes a cohort of tracked GitHub issues or ticket tranches, verifies pairwise tree-disjoint package boundaries, isolates concurrent workers in detached worktrees, spawns user-visible detached OpenCode processes (`opencode run --variant high --auto`), enforces direct leaf execution within package boundaries (prohibiting nested `task` calls to prevent depth limit recursion failures, #12028), and supervises execution through automated harvest and safe git landing.\n\n---\n\n## Six Invariants of Deterministic OpenCode Wave Dispatch\n\n1. **GitHub Issues & Tranche Grounding First**: Every substantive unit of work must be tracked in a GitHub issue and grounded on ticket specifications before worker execution begins. When targeting companion repositories (e.g. `anthony-chaudhary\u002Ffak-private`), explicitly bind candidate issues to the target repo and local ticket tranche (`docs\u002Ftickets\u002F...\u002FINDEX.md`).\n2. **Mandatory Worktree Isolation**: Dispatched workers touching identical Go packages must execute in detached worker worktrees (`--worktree`) or be partitioned into sequential waves. Concurrent edits to shared packages in the same working tree cause compilation cross-contamination and test failures.\n3. **Strict Ban on Loose Scripts (Modular Native Go Dispatch Only)**: Creating new PowerShell (`.ps1`), Bash (`.sh`), or batch (`.bat`) scripts is prohibited. Wave planning, process spawning, and harvest reconciliation must run through first-class Go tooling (`fak issue-orchestrator`).\n4. **High-Reasoning Effort & Auto-Approve Flags**: Dispatched child OpenCode sessions run with `--variant high --auto` (or `--dangerously-skip-permissions`). This activates Gemini 3.8 Flash high reasoning mode for complex debugging, concurrency verification, and kernel implementation.\n5. **Leaf Worker Direct Execution (No Nested Subagents)**: Dispatched sessions act as depth-1 leaf workers focused on assigned lane and boundary paths. Workers execute deliverables directly within package boundaries, author reproduction tests, and run package verification without calling the `task` tool or attempting nested subagent delegation. The coordinator manages wave fan-out; leaf workers execute directly to prevent recursion depth limit failures (`Subagent depth limit reached (1)`, #12028, #12029).\n6. **Automated Closed-Loop Supervision & Harvest**: Waves do not stop at process launch. The coordinator supervises active workers, enforces the 4-state harvest protocol (`StateVerifiedCleared`, `StateResidualReview`, `StateQuietIncomplete`, `StateSpinningStalled`), lands verified worktrees to trunk, and verifies GitHub comment receipts.\n\n---\n\n## Execution Protocol\n\n### Phase 1: Intake & Issue Tranche Ingestion\n\nQuery candidate issues from the target repository and pipe directly into structured JSON:\n\n```bash\n# Ingest candidate issues from target repository (public fak or private companion)\ngh issue list -R anthony-chaudhary\u002Ffak-private --limit 15 --json number,title,labels,body > _scratch\u002Fissues.json\n```\n\nWhen executing a local campaign tranche, reconcile against the ticket index:\n```bash\n# Verify ticket definitions and acceptance criteria\ncat docs\u002Ftickets\u002F\u003Ccampaign>\u002FINDEX.md\n```\n\n### Phase 2: Tree-Disjoint Wave Partitioning & Lease Arbitration\n\nPartition the candidate issue set into ordered, collision-free waves where no two concurrent issues in the same wave touch overlapping packages:\n\n```bash\n# Plan safe waves with pairwise tree-disjoint boundaries and lease verification\ngo run .\u002Fcmd\u002Ffak issue-orchestrator \\\n  --workspace . \\\n  --from-issues _scratch\u002Fissues.json \\\n  --wave-size 4 \\\n  --plan-waves \\\n  --worktree \\\n  --json\n```\n\nVerify that each planned wave respects lane leases (`dos arbitrate`) and isolates high-blast-radius core packages into single-worker waves.\n\n### Phase 3: Native Go Wave Dispatch\n\nLaunch detached OpenCode worker processes for the target wave using first-class native Go tooling:\n\n```bash\n# Dispatch Wave 1 with high reasoning effort and detached worktree isolation\ngo run .\u002Fcmd\u002Ffak issue-orchestrator \\\n  --workspace . \\\n  --from-issues _scratch\u002Fissues.json \\\n  --spawn-opencode \\\n  --spawn-wave 1 \\\n  --wave-size 4 \\\n  --variant high \\\n  --worktree \\\n  --json\n```\n\nFor private platform components in `fak-private`, specify the companion workspace root:\n```bash\ngo run .\u002Fcmd\u002Ffak issue-orchestrator \\\n  --workspace ..\u002Ffak-private \\\n  --from-issues _scratch\u002Fissues.json \\\n  --spawn-opencode \\\n  --spawn-wave 1 \\\n  --variant high \\\n  --worktree \\\n  --json\n```\n\n### Phase 4: Live Supervision & Watchdog Monitoring\n\nAudit running worker processes, session database registrations, and log streams:\n\n```bash\n# Check registered OpenCode database sessions\nopencode session list\n\n# Check running processes and resource utilization\npowershell -Command \"Get-Process -Name 'opencode' | Select-Object Id, ProcessName, CPU, WorkingSet64, StartTime | Format-Table -AutoSize\"\n\n# Inspect real-time execution log of active worker\ncat _scratch\u002Flogs\u002Fopencode-issue-\u003Cissue_number>.err.log\n```\n\n### Phase 5: Automated Closed-Loop Harvest & Safe Git Landing\n\nReconcile wave outcomes against the 4-state harvest protocol:\n\n```bash\n# Execute automated wave harvest and receipt generation\ngo run .\u002Fcmd\u002Ffak issue-orchestrator \\\n  --workspace . \\\n  --harvest \\\n  --json\n```\n\nReconciliation states:\n- `StateVerifiedCleared`: Git commit witnessed on worktree, tests pass green, and 5-gate boundary clean. Auto-land worktree to trunk (`fak-flow land` or `fak sync push`) and close\u002Fcomment on GitHub issue.\n- `StateResidualReview`: Diffs exist but uncorroborated or tests failing. Flag for manual inspection.\n- `StateQuietIncomplete`: No commits or modifications found. Soft re-queue for next wave.\n- `StateSpinningStalled`: Process exceeded turn budget or deadlocked. Terminate PID and release lease.\n\nVerify on-device package tests before closing the wave:\n```bash\ngo test -v .\u002Finternal\u002F\u003Cchanged-pkg>\u002F...\ngo vet .\u002Finternal\u002F\u003Cchanged-pkg>\u002F...\n```\n\n---\n\n## Verification and Witness\n\nVerify this skill definition using the project gates:\n\n```bash\n# 1. Structural admission and anti-slop verification:\npython tools\u002Fskill_slop_scorecard.py .claude\u002Fskills\u002Fopencode-wave\u002FSKILL.md --corpus .claude\u002Fskills\n\n# 2. Frontmatter portability check:\npython tools\u002Fskill_frontmatter_lint.py --check\n\n# 3. Synchronize cross-harness adapter for OpenCode (.agents\u002Fskills\u002Fopencode-wave\u002FSKILL.md):\ngo run .\u002Fcmd\u002Ffak-project-assets sync --json\n\n# 4. Verify zero unexplained parity gaps across harnesses:\ngo run .\u002Fcmd\u002Ffak-project-assets parity --json\n```\n","# \u002Fopencode-wave — High-Effort Deterministic OpenCode Multi-Agent Wave Dispatch\n\nThe campaign coordinator for running multi-issue resolution waves in OpenCode. It takes a cohort of tracked GitHub issues or ticket tranches, verifies pairwise tree-disjoint package boundaries, isolates concurrent workers in detached worktrees, spawns user-visible detached OpenCode processes (`opencode run --variant high --auto`), enforces direct leaf execution within package boundaries (prohibiting nested `task` calls to prevent depth limit recursion failures, #12028), and supervises execution through automated harvest and safe git landing.\n\n---\n\n## Six Invariants of Deterministic OpenCode Wave Dispatch\n\n1. **GitHub Issues & Tranche Grounding First**: Every substantive unit of work must be tracked in a GitHub issue and grounded on ticket specifications before worker execution begins. When targeting companion repositories (e.g. `anthony-chaudhary\u002Ffak-private`), explicitly bind candidate issues to the target repo and local ticket tranche (`docs\u002Ftickets\u002F...\u002FINDEX.md`).\n2. **Mandatory Worktree Isolation**: Dispatched workers touching identical Go packages must execute in detached worker worktrees (`--worktree`) or be partitioned into sequential waves. Concurrent edits to shared packages in the same working tree cause compilation cross-contamination and test failures.\n3. **Strict Ban on Loose Scripts (Modular Native Go Dispatch Only)**: Creating new PowerShell (`.ps1`), Bash (`.sh`), or batch (`.bat`) scripts is prohibited. Wave planning, process spawning, and harvest reconciliation must run through first-class Go tooling (`fak issue-orchestrator`).\n4. **High-Reasoning Effort & Auto-Approve Flags**: Dispatched child OpenCode sessions run with `--variant high --auto` (or `--dangerously-skip-permissions`). This activates Gemini 3.8 Flash high reasoning mode for complex debugging, concurrency verification, and kernel implementation.\n5. **Leaf Worker Direct Execution (No Nested Subagents)**: Dispatched sessions act as depth-1 leaf workers focused on assigned lane and boundary paths. Workers execute deliverables directly within package boundaries, author reproduction tests, and run package verification without calling the `task` tool or attempting nested subagent delegation. The coordinator manages wave fan-out; leaf workers execute directly to prevent recursion depth limit failures (`Subagent depth limit reached (1)`, #12028, #12029).\n6. **Automated Closed-Loop Supervision & Harvest**: Waves do not stop at process launch. The coordinator supervises active workers, enforces the 4-state harvest protocol (`StateVerifiedCleared`, `StateResidualReview`, `StateQuietIncomplete`, `StateSpinningStalled`), lands verified worktrees to trunk, and verifies GitHub comment receipts.\n\n---\n\n## Execution Protocol\n\n### Phase 1: Intake & Issue Tranche Ingestion\n\nQuery candidate issues from the target repository and pipe directly into structured JSON:\n\n```bash\n# Ingest candidate issues from target repository (public fak or private companion)\ngh issue list -R anthony-chaudhary\u002Ffak-private --limit 15 --json number,title,labels,body > _scratch\u002Fissues.json\n```\n\nWhen executing a local campaign tranche, reconcile against the ticket index:\n```bash\n# Verify ticket definitions and acceptance criteria\ncat docs\u002Ftickets\u002F\u003Ccampaign>\u002FINDEX.md\n```\n\n### Phase 2: Tree-Disjoint Wave Partitioning & Lease Arbitration\n\nPartition the candidate issue set into ordered, collision-free waves where no two concurrent issues in the same wave touch overlapping packages:\n\n```bash\n# Plan safe waves with pairwise tree-disjoint boundaries and lease verification\ngo run .\u002Fcmd\u002Ffak issue-orchestrator \\\n  --workspace . \\\n  --from-issues _scratch\u002Fissues.json \\\n  --wave-size 4 \\\n  --plan-waves \\\n  --worktree \\\n  --json\n```\n\nVerify that each planned wave respects lane leases (`dos arbitrate`) and isolates high-blast-radius core packages into single-worker waves.\n\n### Phase 3: Native Go Wave Dispatch\n\nLaunch detached OpenCode worker processes for the target wave using first-class native Go tooling:\n\n```bash\n# Dispatch Wave 1 with high reasoning effort and detached worktree isolation\ngo run .\u002Fcmd\u002Ffak issue-orchestrator \\\n  --workspace . \\\n  --from-issues _scratch\u002Fissues.json \\\n  --spawn-opencode \\\n  --spawn-wave 1 \\\n  --wave-size 4 \\\n  --variant high \\\n  --worktree \\\n  --json\n```\n\nFor private platform components in `fak-private`, specify the companion workspace root:\n```bash\ngo run .\u002Fcmd\u002Ffak issue-orchestrator \\\n  --workspace ..\u002Ffak-private \\\n  --from-issues _scratch\u002Fissues.json \\\n  --spawn-opencode \\\n  --spawn-wave 1 \\\n  --variant high \\\n  --worktree \\\n  --json\n```\n\n### Phase 4: Live Supervision & Watchdog Monitoring\n\nAudit running worker processes, session database registrations, and log streams:\n\n```bash\n# Check registered OpenCode database sessions\nopencode session list\n\n# Check running processes and resource utilization\npowershell -Command \"Get-Process -Name 'opencode' | Select-Object Id, ProcessName, CPU, WorkingSet64, StartTime | Format-Table -AutoSize\"\n\n# Inspect real-time execution log of active worker\ncat _scratch\u002Flogs\u002Fopencode-issue-\u003Cissue_number>.err.log\n```\n\n### Phase 5: Automated Closed-Loop Harvest & Safe Git Landing\n\nReconcile wave outcomes against the 4-state harvest protocol:\n\n```bash\n# Execute automated wave harvest and receipt generation\ngo run .\u002Fcmd\u002Ffak issue-orchestrator \\\n  --workspace . \\\n  --harvest \\\n  --json\n```\n\nReconciliation states:\n- `StateVerifiedCleared`: Git commit witnessed on worktree, tests pass green, and 5-gate boundary clean. Auto-land worktree to trunk (`fak-flow land` or `fak sync push`) and close\u002Fcomment on GitHub issue.\n- `StateResidualReview`: Diffs exist but uncorroborated or tests failing. Flag for manual inspection.\n- `StateQuietIncomplete`: No commits or modifications found. Soft re-queue for next wave.\n- `StateSpinningStalled`: Process exceeded turn budget or deadlocked. Terminate PID and release lease.\n\nVerify on-device package tests before closing the wave:\n```bash\ngo test -v .\u002Finternal\u002F\u003Cchanged-pkg>\u002F...\ngo vet .\u002Finternal\u002F\u003Cchanged-pkg>\u002F...\n```\n\n---\n\n## Verification and Witness\n\nVerify this skill definition using the project gates:\n\n```bash\n# 1. Structural admission and anti-slop verification:\npython tools\u002Fskill_slop_scorecard.py .claude\u002Fskills\u002Fopencode-wave\u002FSKILL.md --corpus .claude\u002Fskills\n\n# 2. Frontmatter portability check:\npython tools\u002Fskill_frontmatter_lint.py --check\n\n# 3. Synchronize cross-harness adapter for OpenCode (.agents\u002Fskills\u002Fopencode-wave\u002FSKILL.md):\ngo run .\u002Fcmd\u002Ffak-project-assets sync --json\n\n# 4. Verify zero unexplained parity gaps across harnesses:\ngo run .\u002Fcmd\u002Ffak-project-assets parity --json\n```\n",{"name":5,"description":7,"allowed-tools":24,"argument-hint":25,"user-invocable":26,"disable-model-invocation":20},"Read, Edit, Write, Grep, Glob, Bash, Task","[--top 10] [--wave-size 4] [--variant high] [--worktree] [--workspace \u003Cpath>] [--from-issues \u003Cfile>] [--harvest]",true,"main","f565b7b7db63e79649f3a02c08108132a86c5d05",".claude\u002Fskills\u002Fopencode-wave",16,"https:\u002F\u002Fgithub.com\u002Fanthony-chaudhary\u002Ffak","https:\u002F\u002Fanthony-chaudhary.github.io\u002Ffak\u002Fshowcase.html","https:\u002F\u002Fgithub.com\u002Fanthony-chaudhary","https:\u002F\u002Flinkedin.com\u002Fin\u002Fanthonychaudhary","2026-11-15 03:06:00",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":26},[38,41,44],{"tool":39,"command":40},"npx","npx skills add anthony-chaudhary\u002Ffak --skill opencode-wave",{"tool":42,"command":43},"bunx","bunx skills add anthony-chaudhary\u002Ffak --skill opencode-wave",{"tool":45,"command":46},"pnpm","pnpm dlx skills add anthony-chaudhary\u002Ffak --skill opencode-wave","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fanthony-chaudhary\u002Ffak\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fopencode-wave\u002FSKILL.md",[],[50,57,66,73,80,85],{"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},"01M78V4059KRP2T88YR6Z05MJQ","serve-receipt","On-demand research protocol + deterministic receipt for serving a Hugging Face model on a concrete GPU target — fan out research lanes (model facts\u002Ffiles, engine support matrix, serving recipes, KV\u002Fstate memory math), compute the tiered feasibility verdict with `fak serve-receipt` (never narrated math), and emit a witnessed launch receipt. Use when asked for an \"initial config\", \"serving receipt\", or \"how do I serve \u003Cmodel> on \u003CGPUs>\".","https:\u002F\u002Fgithub.com\u002Fanthony-chaudhary\u002Ffak\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fserve-receipt\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":61,"licenseFileUrl":62,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":19},"01M77HRGKQV2CJ2K6ZFN4YQ32K","caveman","Ultra-compressed communication mode. Cuts output tokens 65% (measured) by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says \"caveman mode\", \"talk like caveman\", \"use caveman\", \"less tokens\", \"be brief\", or invokes \u002Fcaveman. Also auto-triggers when token efficiency is requested.","MIT","https:\u002F\u002Fgithub.com\u002Fanthony-chaudhary\u002Ffak\u002Fblob\u002Fmain\u002Finternal\u002Fsyspromptmmu\u002Fthird_party\u002Fcaveman\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fanthony-chaudhary\u002Ffak\u002Fblob\u002Fmain\u002Finternal\u002Fsyspromptmmu\u002Fthird_party\u002Fcaveman\u002FSKILL.md",[],[],{"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},"01M77HRAY3Y3BE7APN5PVJF7NV","quality-report","Score the repo quality and emit a report card.","https:\u002F\u002Fgithub.com\u002Fanthony-chaudhary\u002Ffak\u002Fblob\u002Fmain\u002Finternal\u002Fctxplans\u002Ftestdata\u002Frepo\u002F.claude\u002Fskills\u002Fquality-report\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},"01M77HR9807KHAX13DG9PW2K2Q","ctx-overlay","Compact the context window and prune the memory store when a session grows too long.","https:\u002F\u002Fgithub.com\u002Fanthony-chaudhary\u002Ffak\u002Fblob\u002Fmain\u002Finternal\u002Fctxplans\u002Ftestdata\u002Frepo\u002F.claude\u002Fskills\u002Fctx-overlay\u002FSKILL.md",[],[],{"skillNo":81,"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":82,"categories":83,"keywords":84,"publishedAt":19},"01M77HR7XQ9X8F2CY9Z4F2ZEZ3","https:\u002F\u002Fgithub.com\u002Fanthony-chaudhary\u002Ffak\u002Fblob\u002Fmain\u002Finternal\u002Fctxplanlint\u002Ftestdata\u002Frepo\u002F.claude\u002Fskills\u002Fquality-report\u002FSKILL.md",[],[],{"skillNo":86,"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":87,"categories":88,"keywords":89,"publishedAt":19},"01M77HR5BVKCTR1WNJGRBSDNX7","https:\u002F\u002Fgithub.com\u002Fanthony-chaudhary\u002Ffak\u002Fblob\u002Fmain\u002Finternal\u002Fctxplanlint\u002Ftestdata\u002Frepo\u002F.claude\u002Fskills\u002Fctx-overlay\u002FSKILL.md",[],[]]