{
  "schemaVersion": 1,
  "version": 1,
  "author": "Jasmine James",
  "id": "choose-ai-assistance",
  "title": "Choose a step to improve—and a responsibility to protect",
  "summary": "Choose one recurring burden to improve with AI while protecting judgment and useful practice.",
  "inputs": [
    "A recurring workflow and its steps",
    "Who owns the outcome and who currently does each step",
    "Customer outcome or burden to improve",
    "Judgment, relationships or development opportunities to protect"
  ],
  "prompt": "Help me choose a bounded AI-assisted experiment using ownership and leverage.\n\nPREFLIGHT\nVerify which tools are actually available and which source locations they can access. Do not assume that naming an MCP server means it is connected. Local file and terminal tools are valid and do not necessarily use MCP. If required access is missing, ask for an approved export or excerpt; state what cannot be checked or run.\n\nINPUTS\nWorkflow and steps: [fill in]\nOutcome owner / current contributors: [fill in]\nCustomer outcome or recurring burden: [fill in]\nJudgment, relationships or growth to protect: [fill in]\nAvailable baseline evidence: [fill in or state none]\n\nASSESS\nConsider the steps separately, not the whole job. For each, ask: who should own the outcome, and where does continued human attention change it? Identify familiar repetition that may benefit from assistance and work where judgment, relationships, practice or coaching adds value. A step that is routine for one person may be useful development for another. Low leverage does not mean low importance.\n\nOUTPUT\nRecommend one small experiment. Name the step, intended improvement, bounded AI contribution, necessary context, owner and reviewer. Specify what stays with a person, how quality will be checked, a review date and a concrete stop-or-revise condition. Name the customer outcome or development opportunity the change should support. Separate observed facts from assumptions. If no baseline exists, propose a lightweight way to establish one.\n\nSTOP\nReturn a plan for review; do not automate or change the workflow. Do not assume that assistance frees time or that freed time automatically produces a better outcome. Define the evidence that would let us assess each claim.",
  "review": [
    "Is this the right step to assist, given the person and their development needs?",
    "Are ownership and the human review decision clear?",
    "Is the experiment small enough to test and stop?"
  ],
  "evidence": [
    "Compare the same quality, recurring effort and customer-outcome measures over comparable periods.",
    "Record where attention actually went, rather than counting generated output alone.",
    "Review the protected responsibility or development opportunity alongside efficiency."
  ],
  "status": "Adaptable template based on the talk; not the original internal prompt",
  "tooling": {
    "summary": "No MCP is needed when you provide the workflow and baseline. Optional read-only data access can help establish a baseline before an experiment.",
    "noMcpFallback": "Describe the workflow, ownership and evidence you already have. If no baseline is available, define a small manual measurement rather than assuming a benefit.",
    "requirements": [
      {
        "capability": "Read optional baseline evidence",
        "requiredWhen": "Only when the baseline must be retrieved from an external system rather than supplied by the owner.",
        "providers": [
          "An approved reporting/data connector or MCP",
          "Local CSV or exported report tools"
        ],
        "operations": [
          "Read the agreed quality, effort or outcome measures",
          "Check definitions and reporting periods",
          "Record what is observed and what is missing"
        ],
        "access": "Read access limited to the named measures and reporting window. This planning exercise does not need workflow edits, messaging, administration or automation permissions.",
        "manualAlternative": "Provide an approved aggregate export, record a short manual baseline, or state that the baseline is not yet known.",
        "sourceUrls": []
      }
    ]
  },
  "updated": "2026-11-04",
  "value": {
    "id": "choose-ai-assistance",
    "cardSummary": "Break repeated work into steps and choose one small AI experiment. Keep useful learning and ownership visible.",
    "outcome": "Break repeated work into steps and choose one small AI experiment. Keep useful learning and ownership visible.",
    "bridge": {
      "practice": "Softball IQ gave me a place to try capabilities and build the judgment to use them.",
      "work": "The framework emerged as I reflected on repeated preparation, leadership attention and the work my team should own.",
      "relationship": "shared-pattern"
    },
    "humanDecision": "You choose which steps to improve in the context of each person, their responsibility and their growth.",
    "valueToCheck": "Check the intended result and the human effort required, alongside the judgment, ownership and useful learning you protect."
  },
  "firstRun": {
    "id": "choose-ai-assistance",
    "goal": "Choose one step to improve with AI.",
    "steps": [
      {
        "actor": "human",
        "label": "Describe work and responsibility",
        "description": "Supply the steps, people, intended outcome and baseline. Explicitly name the risk interpretation and coaching that should stay with Sam and Mira."
      },
      {
        "actor": "ai",
        "label": "Design a bounded experiment",
        "description": "Assess each step separately. Propose assistance with collecting dated evidence, then define the owner, quality checks, measurement and stopping condition without assuming a time saving."
      },
      {
        "actor": "human",
        "label": "Decide whether to try it",
        "description": "Mira and Sam review the plan. If they choose to run it, they measure human effort and corrections and review after two comparable reports; no workflow changes are made by this draft."
      }
    ],
    "output": {
      "title": "Two-report experiment proposal · draft",
      "sections": [
        {
          "label": "One step to assist",
          "text": "For the next two comparable reports, propose using AI to turn approved ticket excerpts into a draft table: customer, ticket ID, update date and stated status, with a reference for every row. Mark missing or conflicting updates for review. Mira owns the outcome; Sam prepares and checks the evidence table. [A1, A3]"
        },
        {
          "label": "Responsibility and growth to protect",
          "text": "Sam writes the first interpretation of risks before Mira's review. Mira retains prioritization and customer communication. Keep their coaching discussion. The proposed assistance covers evidence preparation, not those decisions. [A1, A2]"
        },
        {
          "label": "Quality, effort and review point",
          "text": "Sam checks every row against its excerpt and checks that all supplied tickets are represented. Record active preparation, review, correction and maintenance minutes for both trial reports. The supplied gathering baseline averages 20 minutes across three reports; total baseline human effort is unknown. Compare like-for-like gathering separately, and do not declare an overall saving from that number alone. [A3] Review after the second report."
        },
        {
          "label": "Stop or revise condition",
          "text": "Stop using the draft table if it invents a status or source; revise the scope if checking it displaces Sam's risk practice or coaching. These are proposed conditions, not observed failures. Whether assistance improves report accuracy or reduces total effort remains unknown. No automation has been started."
        }
      ]
    },
    "review": [
      "Does this help Sam with repetition while preserving risk interpretation, coaching and Mira's ownership?",
      "Are the two-report scope, source checks and stopping conditions clear enough to assess before continuing?"
    ],
    "supplied": [
      {
        "label": "[A1] Fictional weekly status workflow",
        "text": "Mira owns the weekly customer status report. Analyst Sam gathers dated ticket updates, groups them by customer, proposes an interpretation of risks, and reviews the report with Mira. Mira decides priorities and what to tell customers."
      },
      {
        "label": "[A2] Intended improvement and growth to protect",
        "text": "Sam wants less repeated copying and searching, but is learning to recognize delivery risks. Keep Sam's first interpretation and Mira's coaching discussion. The customer outcome to support is an accurate status report; there is no evidence yet that assistance improves it."
      },
      {
        "label": "[A3] Baseline and available evidence",
        "text": "For three comparable weekly reports, Sam recorded 18, 22 and 20 active minutes gathering updates. Review, correction and maintenance time were not measured. Approved excerpts can include ticket IDs, update dates and status text. No live connector access is needed for a first trial."
      }
    ],
    "exampleUrl": null,
    "outputRequirements": [
      "One step to assist",
      "Responsibility and growth to protect",
      "Quality, effort and review point",
      "Stop or revise condition"
    ],
    "updated": "2026-11-04",
    "notice": "All examples use invented sources and results. They are illustrations, not reports from Jasmine’s workplace or measured outcomes.",
    "modelAgnosticNote": "Use the AI tool approved for your work with supplied material. Wording and layout will vary by model; look for the evidence, uncertainty and review decisions shown here. Connections and recurring runs depend on your setup."
  },
  "modelGuidance": {
    "reviewedOn": "2026-10-30",
    "basis": "Suggested fit based on official provider documentation, not our own speed or quality benchmarks. Test on representative examples before relying on a model.",
    "pricingBasis": "USD per one million input and output tokens, using standard first-party API pricing. These are not subscription prices or total workflow costs. Tool calls, reasoning, retries, caching and different tokenizers affect the final bill.",
    "runtimeNote": "A model does not connect your tools or grant access. MCP support, available tools and model selection depend on the agent host and your organization's approved setup.",
    "catalogUrl": "https://jasminejames.dev/leading-deeper/model-guidance.json",
    "start": {
      "modelId": "gpt-6-luna",
      "effort": "low",
      "why": "An economical first pass through a structured worksheet; keep the experiment small and the output short."
    },
    "escalate": {
      "modelId": "gpt-6.1-sol",
      "when": "Ownership, growth opportunities and competing constraints need a more careful examination."
    },
    "alternatives": [
      {
        "modelId": "claude-haiku-5-5",
        "when": "An alternative for the first-pass plan in your approved Claude-based setup."
      }
    ],
    "humanBoundary": "The leader makes the choice. The model helps examine the tradeoffs; it does not decide whose development or responsibility to remove.",
    "models": {
      "gpt-6-luna": {
        "name": "GPT-6 Luna",
        "provider": "OpenAI",
        "apiId": "gpt-6-luna",
        "context": "1.05 million tokens",
        "providerPositioning": "Focused, cost-sensitive, high-volume work.",
        "prices": [
          {
            "label": "Standard API",
            "inputUsdPerMillion": 0.1,
            "outputUsdPerMillion": 0.5,
            "condition": ""
          }
        ],
        "modelSourceUrl": "https://developers.openai.com/api/docs/models/gpt-6-luna",
        "pricingSourceUrl": "https://developers.openai.com/api/docs/pricing"
      },
      "gpt-6.1-sol": {
        "name": "GPT-6.1 Sol",
        "provider": "OpenAI",
        "apiId": "gpt-6.1-sol",
        "context": "1.05 million tokens",
        "providerPositioning": "Complex coding and professional work with a balance of intelligence and cost.",
        "prices": [
          {
            "label": "Standard API",
            "inputUsdPerMillion": 2,
            "outputUsdPerMillion": 10,
            "condition": ""
          }
        ],
        "modelSourceUrl": "https://developers.openai.com/api/docs/models/gpt-6.1-sol",
        "pricingSourceUrl": "https://developers.openai.com/api/docs/pricing"
      },
      "claude-haiku-5-5": {
        "name": "Claude Haiku 5.5",
        "provider": "Anthropic",
        "apiId": "claude-haiku-5-5",
        "context": "1 million tokens",
        "providerPositioning": "High-volume, latency-sensitive classification, extraction and routing.",
        "prices": [
          {
            "label": "Standard API",
            "inputUsdPerMillion": 0.1,
            "outputUsdPerMillion": 0.5,
            "condition": "For requests with prompts up to 100,000 tokens."
          },
          {
            "label": "Standard API",
            "inputUsdPerMillion": 0.5,
            "outputUsdPerMillion": 2.5,
            "condition": "For requests with prompts over 100,000 tokens; the higher rates apply to that request."
          }
        ],
        "modelSourceUrl": "https://platform.claude.com/docs/en/models/haiku-5-5/overview",
        "pricingSourceUrl": "https://platform.claude.com/docs/en/about-claude/pricing"
      }
    }
  },
  "skillUrl": "https://jasminejames.dev/leading-deeper/skills/find-low-leverage-workflows/SKILL.md",
  "repositorySourceUrl": "https://github.com/jasminelsjames/leading-deeper-ai-skills/blob/e1d2e65d3922c5e376fc0c6e8cfbb2f85722b1a7/skills/find-low-leverage-workflows/SKILL.md",
  "repositoryRawUrl": "https://raw.githubusercontent.com/jasminelsjames/leading-deeper-ai-skills/e1d2e65d3922c5e376fc0c6e8cfbb2f85722b1a7/skills/find-low-leverage-workflows/SKILL.md",
  "skillRequest": "Read this GitHub agent skill and use it to help me apply the AI Leverage Framework:\nhttps://raw.githubusercontent.com/jasminelsjames/leading-deeper-ai-skills/e1d2e65d3922c5e376fc0c6e8cfbb2f85722b1a7/skills/find-low-leverage-workflows/SKILL.md\n\nRecurring task: [what happens and how often]\nCurrent steps: [how you do it today]\nPeople: [who does it and who owns the outcome]\nKeep human: [judgment, relationships or learning to protect]\nTime and example: [if available]\n\nStart with my recent work. Identify repeated steps I could improve, what judgment or learning to protect and one small experiment to try. Work from the examples I provide or read access I authorize. Propose what to make room for and how to assess whether the experiment helped.",
  "sampleRequest": "Read this GitHub agent skill and use it to help me apply the AI Leverage Framework:\nhttps://raw.githubusercontent.com/jasminelsjames/leading-deeper-ai-skills/e1d2e65d3922c5e376fc0c6e8cfbb2f85722b1a7/skills/find-low-leverage-workflows/SKILL.md\n\nTry this fictional example. All people, records and measurements below are invented. Use only the supplied context; no connectors, account access or additional records are available.\n\nGoal: One repeated task to improve, with a small AI experiment to try.\nYou prepare the draft; I review it. Keep any named decision owner in the supplied material responsible for their decision.\n\nSupplied context:\n[A1] Fictional weekly status workflow\nMira owns the weekly customer status report. Analyst Sam gathers dated ticket updates, groups them by customer, proposes an interpretation of risks, and reviews the report with Mira. Mira decides priorities and what to tell customers.\n\n[A2] Intended improvement and growth to protect\nSam wants less repeated copying and searching, but is learning to recognize delivery risks. Keep Sam's first interpretation and Mira's coaching discussion. The customer outcome to support is an accurate status report; there is no evidence yet that assistance improves it.\n\n[A3] Baseline and available evidence\nFor three comparable weekly reports, Sam recorded 18, 22 and 20 active minutes gathering updates. Review, correction and maintenance time were not measured. Approved excerpts can include ticket IDs, update dates and status text. No live connector access is needed for a first trial.\n\nReturn:\n- One step to assist\n- Responsibility and growth to protect\n- Quality, effort and review point\n- Stop or revise condition\n\nStop after returning the private draft in this chat. Keep source labels, assumptions and unknowns visible. Do not send messages, change files or tasks, publish, connect tools or create a schedule. If you cannot read the linked skill, say so rather than claiming to have used it.",
  "feedbackUrl": "https://github.com/jasminelsjames/leading-deeper-ai-skills/issues/new?template=skill-feedback.md&title=Feedback%3A+Choose+Where+Attention+Belongs&body=**Which+skill+did+you+try%3F**%0A%0AChoose+Where+Attention+Belongs+%28find-low-leverage-workflows%29%0A%0A**What+did+you+use+it+for%2C+and+what+result+did+you+get%3F**%0A%0A**What+did+you+have+to+correct%2C+and+what+would+make+the+next+run+better%3F**%0A%0A**Agent+and+model%2C+if+known%3A**%0A%0AOptional%3A+a+short+fictional+or+sanitized+example.+Do+not+include+private+workplace+content%2C+credentials+or+personal+information+about+others.",
  "frameworkArtifactsGuide": {
    "id": "framework-artifacts",
    "title": "What each quadrant leaves behind",
    "summary": "Turn repeated work into reusable AI workflows. Make decisions, ownership and growth visible.",
    "htmlUrl": "https://jasminejames.dev/leading-deeper/choose-ai-assistance/#framework-artifacts",
    "markdownUrl": "https://jasminejames.dev/leading-deeper/framework-artifacts.md",
    "jsonUrl": "https://jasminejames.dev/leading-deeper/framework-artifacts.json"
  },
  "htmlUrl": "https://jasminejames.dev/leading-deeper/choose-ai-assistance",
  "markdownUrl": "https://jasminejames.dev/leading-deeper/choose-ai-assistance.md",
  "jsonUrl": "https://jasminejames.dev/leading-deeper/choose-ai-assistance.json",
  "textUrl": "https://jasminejames.dev/leading-deeper/prompts/choose-ai-assistance.txt",
  "requiresHumanReview": true
}
