[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79BFEGRXNHXA1CDBK2450JB":3,"skill-same:01M79BFEGRXNHXA1CDBK2450JB":52},{"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":30,"commitSha":31,"skillDirectory":32,"forksCount":14,"repositoryHtmlUrl":33,"repositoryHomepageUrl":6,"ownerHtmlUrl":34,"ownerBlogUrl":35,"ownerDescription":36,"updatedAt":37,"install":38,"resources":51},"01M79BFEGRXNHXA1CDBK2450JB","cost-tracking",null,"Per-task cost accounting protocol. Logs token usage and USD cost to the SQLite database for every task, enabling budget control, model comparison, and cost efficiency analysis by the Harness Critic and Evolution Manager.","sahil-kaushik1\u002FAuditing-Workflow","sahil-kaushik1","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F143880923?v=4","01M79BEEK3947RQYD3WC00K9W3","Sahil Kaushik","01M79BEEKVHPFPTNBDB0GFXBHE",0,"MIT","https:\u002F\u002Fgithub.com\u002Fsahil-kaushik1\u002FAuditing-Workflow\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcost-tracking\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: cost-tracking\ndescription: >\n  Per-task cost accounting protocol. Logs token usage and USD cost to the\n  SQLite database for every task, enabling budget control, model comparison,\n  and cost efficiency analysis by the Harness Critic and Evolution Manager.\ntrigger: After every task completes (any verdict). Mandatory in all modes.\nagents: [orchestrator, evolution-manager, harness-critic]\n---\n\n# Cost Tracking Skill\n\n## Why Track Cost\n\nWithout cost data:\n- Cannot enforce budget_tokens stop rules\n- Cannot compare model efficiency (o3 vs Opus vs Sonnet)\n- Cannot calculate ROI per finding\n- Cannot optimize scanner mode model assignments\n- Harness Critic has no efficiency data\n\n## What to Track Per Task\n\nAfter each task completes (any verdict: CONFIRMED, REFUTED, INCONCLUSIVE, KILLED):\n\n```\ntask_id:          string\ntokens_input:     int   (tokens in the task prompt + context)\ntokens_output:    int   (tokens in the reasoning output)\ntokens_total:     int   (input + output)\ncost_usd:         float (calculated from model-config.yml pricing)\nmodel_used:       string (actual model name, not alias)\nphase:            string (from task.phase)\nverdict:          string\nover_budget:      bool  (tokens_total > task.budget_tokens)\n```\n\n## Cost Calculation\n\nFrom `model-config.yml`, read the active provider's pricing:\n\n```yaml\n# model-config.yml pricing section (per 1M tokens)\nproviders:\n  anthropic:\n    pricing:\n      claude-opus-4:\n        input_per_1m:  15.00\n        output_per_1m: 75.00\n      claude-sonnet-4:\n        input_per_1m:  3.00\n        output_per_1m: 15.00\n      claude-haiku-3:\n        input_per_1m:  0.25\n        output_per_1m: 1.25\n  openai:\n    pricing:\n      o3:\n        input_per_1m:  10.00\n        output_per_1m: 40.00\n      gpt-4.1:\n        input_per_1m:  2.00\n        output_per_1m: 8.00\n```\n\n```bash\n# Calculate cost after each task\ncost_usd=$(echo \"scale=6; ($input_tokens * $input_price + $output_tokens * $output_price) \u002F 1000000\" | bc)\n```\n\n## DB Write Protocol\n\nAfter each task, the Orchestrator writes to the DB:\n\n```sql\nUPDATE tasks\nSET tokens_used = {total_tokens},\n    cost_usd = {cost_usd},\n    status = '{verdict}',\n    completed_at = '{timestamp}',\n    model_reasoner = '{model_name}'\nWHERE task_id = '{task_id}';\n\n-- Also update the run total\nUPDATE runs\nSET total_cost_usd = total_cost_usd + {cost_usd},\n    total_tokens = total_tokens + {total_tokens}\nWHERE run_id = '{run_id}';\n```\n\n## Budget Enforcement\n\nThe Orchestrator enforces the `budget_tokens` field from each task:\n\n```\nBefore routing to Deep Reasoner:\n  - Set context limit = task.budget_tokens\n  - Include in Reasoner prompt: \"STOP if you have used {budget_tokens} tokens.\n    Apply your stop_rule and output current status.\"\n\nAfter Reasoner completes:\n  - If tokens_used > budget_tokens × 1.2 (20% overrun): flag as over_budget\n  - Log to Harness Critic: task over-budget by {N} tokens\n  - This is a harness discipline failure, not a finding failure\n```\n\n## Cost Efficiency Metrics (for Harness Critic)\n\n```sql\n-- Cost per confirmed finding\nSELECT\n  run_id,\n  SUM(cost_usd) as total_cost,\n  COUNT(CASE WHEN verdict = 'CONFIRMED' THEN 1 END) as confirmed,\n  ROUND(SUM(cost_usd) \u002F NULLIF(COUNT(CASE WHEN verdict = 'CONFIRMED' THEN 1 END), 0), 2)\n    as cost_per_confirmed_finding\nFROM tasks\nGROUP BY run_id;\n\n-- Most expensive tasks (potential over-analysis)\nSELECT task_id, target, cost_usd, tokens_used, budget_tokens, verdict\nFROM tasks\nORDER BY cost_usd DESC\nLIMIT 10;\n\n-- Model cost comparison across runs\nSELECT model_reasoner, AVG(cost_usd) as avg_cost, COUNT(*) as task_count,\n       COUNT(CASE WHEN verdict = 'CONFIRMED' THEN 1 END) as confirmed\nFROM tasks\nGROUP BY model_reasoner;\n\n-- Over-budget rate per mode\nSELECT r.mode, COUNT(*) as tasks,\n       SUM(CASE WHEN t.tokens_used > t.budget_tokens THEN 1 ELSE 0 END) as over_budget,\n       ROUND(100.0 * SUM(CASE WHEN t.tokens_used > t.budget_tokens THEN 1 ELSE 0 END) \u002F COUNT(*), 1)\n         as over_budget_pct\nFROM tasks t JOIN runs r ON t.run_id = r.run_id\nGROUP BY r.mode;\n```\n\n## Dashboard Cost Section\n\nThe Orchestrator updates `findings\u002FDASHBOARD.md` cost section after each task:\n\n```markdown\n## Budget Status\n- Total cost so far: ${total_cost_usd}\n- Budget remaining: ${budget_remaining} ({pct}%)\n- Cost per resolved task: ${avg_cost_per_task}\n- Cost per confirmed finding: ${cost_per_finding}\n- Over-budget tasks: {N} ({pct}%)\n```\n","# Cost Tracking Skill\n\n## Why Track Cost\n\nWithout cost data:\n- Cannot enforce budget_tokens stop rules\n- Cannot compare model efficiency (o3 vs Opus vs Sonnet)\n- Cannot calculate ROI per finding\n- Cannot optimize scanner mode model assignments\n- Harness Critic has no efficiency data\n\n## What to Track Per Task\n\nAfter each task completes (any verdict: CONFIRMED, REFUTED, INCONCLUSIVE, KILLED):\n\n```\ntask_id:          string\ntokens_input:     int   (tokens in the task prompt + context)\ntokens_output:    int   (tokens in the reasoning output)\ntokens_total:     int   (input + output)\ncost_usd:         float (calculated from model-config.yml pricing)\nmodel_used:       string (actual model name, not alias)\nphase:            string (from task.phase)\nverdict:          string\nover_budget:      bool  (tokens_total > task.budget_tokens)\n```\n\n## Cost Calculation\n\nFrom `model-config.yml`, read the active provider's pricing:\n\n```yaml\n# model-config.yml pricing section (per 1M tokens)\nproviders:\n  anthropic:\n    pricing:\n      claude-opus-4:\n        input_per_1m:  15.00\n        output_per_1m: 75.00\n      claude-sonnet-4:\n        input_per_1m:  3.00\n        output_per_1m: 15.00\n      claude-haiku-3:\n        input_per_1m:  0.25\n        output_per_1m: 1.25\n  openai:\n    pricing:\n      o3:\n        input_per_1m:  10.00\n        output_per_1m: 40.00\n      gpt-4.1:\n        input_per_1m:  2.00\n        output_per_1m: 8.00\n```\n\n```bash\n# Calculate cost after each task\ncost_usd=$(echo \"scale=6; ($input_tokens * $input_price + $output_tokens * $output_price) \u002F 1000000\" | bc)\n```\n\n## DB Write Protocol\n\nAfter each task, the Orchestrator writes to the DB:\n\n```sql\nUPDATE tasks\nSET tokens_used = {total_tokens},\n    cost_usd = {cost_usd},\n    status = '{verdict}',\n    completed_at = '{timestamp}',\n    model_reasoner = '{model_name}'\nWHERE task_id = '{task_id}';\n\n-- Also update the run total\nUPDATE runs\nSET total_cost_usd = total_cost_usd + {cost_usd},\n    total_tokens = total_tokens + {total_tokens}\nWHERE run_id = '{run_id}';\n```\n\n## Budget Enforcement\n\nThe Orchestrator enforces the `budget_tokens` field from each task:\n\n```\nBefore routing to Deep Reasoner:\n  - Set context limit = task.budget_tokens\n  - Include in Reasoner prompt: \"STOP if you have used {budget_tokens} tokens.\n    Apply your stop_rule and output current status.\"\n\nAfter Reasoner completes:\n  - If tokens_used > budget_tokens × 1.2 (20% overrun): flag as over_budget\n  - Log to Harness Critic: task over-budget by {N} tokens\n  - This is a harness discipline failure, not a finding failure\n```\n\n## Cost Efficiency Metrics (for Harness Critic)\n\n```sql\n-- Cost per confirmed finding\nSELECT\n  run_id,\n  SUM(cost_usd) as total_cost,\n  COUNT(CASE WHEN verdict = 'CONFIRMED' THEN 1 END) as confirmed,\n  ROUND(SUM(cost_usd) \u002F NULLIF(COUNT(CASE WHEN verdict = 'CONFIRMED' THEN 1 END), 0), 2)\n    as cost_per_confirmed_finding\nFROM tasks\nGROUP BY run_id;\n\n-- Most expensive tasks (potential over-analysis)\nSELECT task_id, target, cost_usd, tokens_used, budget_tokens, verdict\nFROM tasks\nORDER BY cost_usd DESC\nLIMIT 10;\n\n-- Model cost comparison across runs\nSELECT model_reasoner, AVG(cost_usd) as avg_cost, COUNT(*) as task_count,\n       COUNT(CASE WHEN verdict = 'CONFIRMED' THEN 1 END) as confirmed\nFROM tasks\nGROUP BY model_reasoner;\n\n-- Over-budget rate per mode\nSELECT r.mode, COUNT(*) as tasks,\n       SUM(CASE WHEN t.tokens_used > t.budget_tokens THEN 1 ELSE 0 END) as over_budget,\n       ROUND(100.0 * SUM(CASE WHEN t.tokens_used > t.budget_tokens THEN 1 ELSE 0 END) \u002F COUNT(*), 1)\n         as over_budget_pct\nFROM tasks t JOIN runs r ON t.run_id = r.run_id\nGROUP BY r.mode;\n```\n\n## Dashboard Cost Section\n\nThe Orchestrator updates `findings\u002FDASHBOARD.md` cost section after each task:\n\n```markdown\n## Budget Status\n- Total cost so far: ${total_cost_usd}\n- Budget remaining: ${budget_remaining} ({pct}%)\n- Cost per resolved task: ${avg_cost_per_task}\n- Cost per confirmed finding: ${cost_per_finding}\n- Over-budget tasks: {N} ({pct}%)\n```\n",{"name":5,"agents":24,"trigger":28,"description":29},[25,26,27],"orchestrator","evolution-manager","harness-critic","After every task completes (any verdict). Mandatory in all modes.","Per-task cost accounting protocol. Logs token usage and USD cost to the SQLite database for every task, enabling budget control, model comparison, and cost efficiency analysis by the Harness Critic and Evolution Manager.\n","main","39a25d63191d67efe43cdce5b7507069e2307f0f",".agents\u002Fskills\u002Fcost-tracking","https:\u002F\u002Fgithub.com\u002Fsahil-kaushik1\u002FAuditing-Workflow","https:\u002F\u002Fgithub.com\u002Fsahil-kaushik1","https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1Te4TcrpgcN4j8DsYqVdjPl4QXVbs5XXo\u002Fview?usp=drive_link","Web3 security researcher \u002F developer","2026-10-18 15:17:21",{"source":8,"skillName":5,"commands":39,"prompt":49,"zipAvailable":50},[40,43,46],{"tool":41,"command":42},"npx","npx skills add sahil-kaushik1\u002FAuditing-Workflow --skill cost-tracking",{"tool":44,"command":45},"bunx","bunx skills add sahil-kaushik1\u002FAuditing-Workflow --skill cost-tracking",{"tool":47,"command":48},"pnpm","pnpm dlx skills add sahil-kaushik1\u002FAuditing-Workflow --skill cost-tracking","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fsahil-kaushik1\u002FAuditing-Workflow\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcost-tracking\u002FSKILL.md",true,[],[53,59,65,72,79,86],{"skillNo":54,"name":55,"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":56,"categories":57,"keywords":58,"publishedAt":19},"01M79BSEAG0SS9CC2EN1S51K2X","stripe-patterns","https:\u002F\u002Fgithub.com\u002Fsahil-kaushik1\u002FAuditing-Workflow\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fsync\u002Frendertrust\u002F.claude\u002Fskills\u002Fstripe-patterns\u002FSKILL.md",[],[],{"skillNo":60,"name":61,"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":62,"categories":63,"keywords":64,"publishedAt":19},"01M79BSBAZXQBWCPKYJP38A35B","payment-patterns","https:\u002F\u002Fgithub.com\u002Fsahil-kaushik1\u002FAuditing-Workflow\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fsync\u002Fkeryk-ai\u002F.claude\u002Fskills\u002Fpayment-patterns\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":19},"01M79BS8PKZXFGBNQBS2GHNXZW","testing-patterns","Testing patterns for Jest and Playwright. Use when writing tests, setting up test fixtures, or validating RLS enforcement. Routes to existing test conventions.","https:\u002F\u002Fgithub.com\u002Fsahil-kaushik1\u002FAuditing-Workflow\u002Fblob\u002Fmain\u002F.gemini\u002Fskills\u002Ftesting-patterns\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":19},"01M79BS3PR92A4QNEJAFNDKB2G","team-coordination","Agent Teams orchestration patterns for multi-agent SAFe workflows. Use when spawning agent teams, coordinating teammates, enforcing quality gates via task dependencies, or orchestrating the 11-agent SAFe pipeline. Covers team creation, messaging, shared task lists, and SAFe gate enforcement. Do NOT use for single-agent or simple subagent workflows.","https:\u002F\u002Fgithub.com\u002Fsahil-kaushik1\u002FAuditing-Workflow\u002Fblob\u002Fmain\u002F.gemini\u002Fskills\u002Fteam-coordination\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":19},"01M79BRZ422QD8D5QHQMXAVPWZ","task-compilation","Compile broad audit goals into structured proof obligations with 9 mandatory fields. Used after decomposition by the Threat Modeler agent.","https:\u002F\u002Fgithub.com\u002Fsahil-kaushik1\u002FAuditing-Workflow\u002Fblob\u002Fmain\u002F.gemini\u002Fskills\u002Ftask-compilation\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":19},"01M79BRV2MK2AJZBNSE2BETP28","storage-layout-analysis","Proxy patterns, slot collisions, upgrade safety. Used during decomposition to identify storage-related vulnerabilities in upgradeable contracts.","https:\u002F\u002Fgithub.com\u002Fsahil-kaushik1\u002FAuditing-Workflow\u002Fblob\u002Fmain\u002F.gemini\u002Fskills\u002Fstorage-layout-analysis\u002FSKILL.md",[],[]]