ai-product-managers-pack

Plugin ai-product-managers-pack from marketplace 90n-skills in polar-bear-org/claude-skills. Commands read from the repo's manifest — 17 days ago.

Skills (36)

• aipm-agent-spec

  Writes an Agent Spec with the goal and done-state, allowed tools and permissions, stop conditions and a spend budget per session, the approval points carried over from the human-in-the-loop design, and the list of actions the agent must never take alone. Use for "run aipm-agent-spec", "agent spec",

• aipm-ai-failure-modes

  Builds an AI Failure Modes Map with the failure list per output type, the cost of a false yes and a false no to the user, severity and detection ratings, a graceful failure per mode and a tolerable error rate left blank for a named person. Use for "run aipm-ai-failure-modes", "AI FMEA", "which AI mi

• aipm-ai-impact-assessment

  Drafts an AI Impact Assessment covering who is affected and how, intended use and foreseeable misuse, data used, human oversight, mitigations and residual risk with a named owner, with every legal point marked as a question for an adviser. Use for "run aipm-ai-impact-assessment", "AI impact assessme

• aipm-ai-incident-response

  Writes an AI Incident Response Plan for a wrong answer or action that reached customers, with severity levels for AI errors, containment (fallback, switch off, human takeover), a customer correction, an evidence timeline, the fix into the golden dataset and a blameless review. Use for "run aipm-ai-i

• aipm-ai-launch-checklist

  Builds an AI Launch Checklist with an owner, a status and an evidence link per line, covering evals passed as run, guardrails on, fallback tested, monitoring and alerts, feedback capture live, support briefed, disclosure in place and rollback rehearsed. Use for "run aipm-ai-launch-checklist", "AI la

• aipm-ai-model-selection

  Builds an AI Model Selection Plan with a side-by-side test plan on 20 to 50 real inputs, a quality, latency and cost table per model tier, a pick per task with the evidence row behind it and a re-check date. Use for "run aipm-ai-model-selection", "which model tier should we use", "justify the model

• aipm-ai-prd

  Drafts an AI PRD with the problem and its evidence, the inputs the model sees, the outputs, a three-line behavior summary, failure handling and fallbacks, data needs, a release bar with blank targets and open questions with owners. Use for "run aipm-ai-prd", "write a PRD for an AI feature", "AI feat

• aipm-ai-prototype-brief

  Writes an AI Prototype Brief with what the demo tests, a pass line set in advance, a cherry-picked input check, what the demo does not prove and a gap-to-production list with an owner per line. Use for "run aipm-ai-prototype-brief", "leadership thinks the demo is the product", "AI demo to production

• aipm-ai-risk-register

  Builds an AI Risk Register with each risk worded as cause, event and effect, mapped to the NIST Map, Measure and Manage functions, with an owner, an early-warning trigger, a response and the regulatory questions passed on to legal. Use for "run aipm-ai-risk-register", "AI risk register", "NIST AI RM

• aipm-ai-success-criteria

  Writes AI Success Criteria for a model feature, one specific and measurable criterion per dimension (task fidelity, consistency, tone, privacy, context use, latency, cost), with today's baseline, a target left blank for a named person and how each will be measured. Use for "run aipm-ai-success-crite

• aipm-ai-use-case-canvas

  Fills an AI Use Case Canvas for one decision or task, with the canvas fields, a simpler-fix check, a data check and a verdict a named person decides. Use for "run aipm-ai-use-case-canvas", "AI canvas", "do we even need AI for this", "leadership wants an AI feature", "is this a good AI use case", "AI

• aipm-ai-ux-review

  Runs an AI UX Review that checks the 18 human-AI interaction guidelines by phase, records what the AI disclosure shows, traces the correction and human handoff paths and lists fixes in the order you set. Use for "run aipm-ai-ux-review", "AI UX review", "human-AI interaction guidelines", "users keep

• aipm-behavior-contract

  Writes an AI Behavior Contract of must, must never and when-unsure lines, each as a Given-When-Then scenario with a pass rate over repeated runs set by a named person, a guardrail list for what the prompt alone cannot enforce, and a tag to the eval that checks each line. Use for "run aipm-behavior-c

• aipm-context-engineering

  Builds a Context Engineering Brief with a register of knowledge sources and their owners, freshness rules, a leave-out list, retrieval spot checks on real questions and a context budget. Use for "run aipm-context-engineering", "context engineering", "which documents should the model read", "the AI a

• aipm-data-privacy-brief

  Drafts an AI Data Privacy Brief with a data flow map from user to model to logs, the personal data in prompts and context with minimisation options, retention and processing region per hop, the vendor terms to read and the questions for legal with the facts attached. Use for "run aipm-data-privacy-b

• aipm-error-analysis

  Reads 50 to 100 real AI outputs with you and turns them into an error analysis log with open codes, failure types grouped by axial coding, counts per type and the three to fix first. Use for "run aipm-error-analysis", "error analysis", "what is actually going wrong with our AI", "read our transcript

• aipm-eval-plan

  Turns the AI success criteria into an eval plan with capability and regression suites, when each runs, what blocks a release and the named owner of quality. Use for "run aipm-eval-plan", "eval plan", "AI eval strategy", "who owns quality", "when should evals run", "regression suite", "capability eva

• aipm-eval-rubric

  Writes an eval rubric with one binary pass or fail check per failure type, the grader for each check (code, model or human), real pass and fail examples, and which checks block a release. Use for "run aipm-eval-rubric", "eval rubric", "pass fail criteria for our AI", "how should we grade outputs", "

• aipm-exec-brief

  Writes a one-page AI Exec Brief, bottom line first, covering what the feature does, how often it is wrong by severity, what a wrong answer costs, how it is caught, run cost and the decision asked. Use for "run aipm-exec-brief", "how accurate is it", "explain the AI to leadership", "one-pager for exe

• aipm-feature-card

  Compiles an AI Feature Card with intended and out-of-scope use, data used, eval results as run with dates, known limits, disclosures, an owner and a review date. Use for "run aipm-feature-card", "model card for our feature", "what does the AI feature do and where does it fail", "AI fact sheet for sa

• aipm-feedback-signals

  Designs an AI Feedback Signal Map with the signals an AI feature should log (accept, edit, retry, rephrase, abandon, escalate, rating), what each likely means, the log spec, a weekly readout and the triggers that send a conversation to review. Use for "run aipm-feedback-signals", "users never rate o

• aipm-golden-dataset

  Builds a golden dataset of 20 to 50 real cases with expected outcomes, coverage by failure type and edge case, source and version per case, and add-and-retire rules. Use for "run aipm-golden-dataset", "golden dataset", "build an eval set", "test cases for our AI feature", "did the new prompt actuall

• aipm-human-in-the-loop

  Designs Human-in-the-Loop for an AI feature or agent, naming which steps a person approves, edits or takes over, the trigger for each (uncertainty, risk, amount, user request), the handoff message and context passed, a response time staff can actually meet and what is logged. Use for "run aipm-human

• aipm-launch-gates

  Drafts AI Launch Gates with release stages (internal, small share, wider, all), the eval and live numbers that must hold at each, kill criteria set in advance, the rollback path and the one named person who signs each gate. Use for "run aipm-launch-gates", "AI launch gates", "staged rollout for an A

• aipm-legal-questions

  Prepares Questions for Legal, grouped by topic (data, disclosure, liability, output ownership, AI Act category, vendor terms, records), each with the facts legal needs attached and the decision it unblocks, and never answers a legal question. Use for "run aipm-legal-questions", "questions for legal"

• aipm-llm-judge

  Writes an LLM-as-a-judge prompt for one failure type and the check that proves it, a labelled agreement set, agreement with human labels on a held-out split, and bias checks for position, length and self-preference. Use for "run aipm-llm-judge", "LLM as a judge", "judge prompt", "automate our eval g

• aipm-model-migration

  Writes a Model Migration Plan for moving off a model with a retirement date, with the deadline from the deprecation notice, breaking changes to check, an eval rerun per failure type, the cost difference, behaviour diffs to read and a rollout with fallback. Use for "run aipm-model-migration", "our mo

• aipm-prompt-regression-test

  Runs a Prompt Regression Test on a prompt, context or setting change, with a change note, the cases to re-run, before and after results per failure type and a ship or hold recommendation for a named person. Use for "run aipm-prompt-regression-test", "did this prompt change break anything", "regressi

• aipm-quality-review

  Runs a Weekly AI Quality Review on live transcripts, with a sample plan, review notes against the failure taxonomy, new failure types, cases added to the golden dataset and decisions with owners. Use for "run aipm-quality-review", "weekly AI quality review", "read our AI transcripts", "it worked for

• aipm-red-team-plan

  Drafts an AI Red Teaming Plan with attack cases by risk class (prompt injection, data leak, excessive agency, system prompt leak, misinformation, runaway cost), who runs each one, a pass line per class and the fixes owed before launch. Use for "run aipm-red-team-plan", "red team our AI feature", "pr

• aipm-synthetic-test-data

  Generates synthetic test cases only for the gaps a real golden dataset cannot cover, each marked synthetic, checked by a person against real cases and reported apart from the headline score. Use for "run aipm-synthetic-test-data", "synthetic test data", "generate edge cases", "we do not have enough

• aipm-system-prompt

  Drafts a System Prompt Brief with the role, audience, task, rules, examples, output format, refusals and handoffs, plus a line-by-line critique of the current prompt against the behavior contract. Use for "run aipm-system-prompt", "write the system prompt", "clean up our prompt", "system prompt temp

• aipm-tool-descriptions

  Writes Tool Descriptions for an agent with each tool's name, purpose, inputs, outputs and error messages, a namespacing scheme, an overlap check between tools and three realistic test tasks. Use for "run aipm-tool-descriptions", "write tool descriptions", "the agent picks the wrong tool", "tool defi

• aipm-unit-economics

  Builds an AI Unit Economics Model with cost per call, per task and per successful outcome (retries and context included), a heavy-user scenario, caching and batch levers and a comparison with the current way of doing the task. Use for "run aipm-unit-economics", "cost per task", "cost per successful

• aipm-usage-and-pricing-test

  Runs an AI Usage and Pricing Test that measures calls, tokens and retries per task on a small set of real tasks, estimates monthly usage at light, normal and heavy use and tests price options against those estimates, as an estimate to test and never pricing advice. Use for "run aipm-usage-and-pricin

• aipm-workflow-or-agent

  Writes a Workflow or Agent Decision comparing the candidate patterns from a single call to an autonomous agent, with cost, latency and failure trade-offs, the simplest pattern that passes and the trigger that would justify the next step up. Use for "run aipm-workflow-or-agent", "do we need an agent"

Similar plugins