More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Job Card: Quote-Backed Jobs-to-be-Done for Every PRD

    1

    Nothing ties the claimed job back to anything a buyer said, so a team can spend weeks on demand that was never there.

    Free

    Security scanned
    Job Card: Quote-Backed Jobs-to-be-Done for Every PRD

    Job Card: Quote-Backed Jobs-to-be-Done for Every PRD

    Example session with this skill installed

    Here's the transcript of my call with a clinic office manager who just signed a pilot. Extract the job card for our scheduling product.

    • Read your context and instructions
    • Compiled the job card

    The agent creates jobs/cover-a-sick-call.md from the template and fills each section (struggle, alternatives, trigger, desired outcomes, anxieties, inertia, stakeholders, commitment signals) with verbatim quotes, each tagged with a timestamp and evidence level. It skips quotes from your own sales rep, adds vague words like "easier" to a list of questions for the next call, searches the transcript to confirm every quote, writes the job statement from the buyer's verbs, and runs the gate: job exists FAIL (4 quotes from 1 source, needs 2 sources), demand exists PASS (the signed pilot).

    Connects securely to your tools. The creator never sees your data.

    About this skill

    About this skill

    Product specs, positioning, and outreach often start from a persona someone remembers or a model imagined. Nothing ties the claimed job back to anything a buyer said, so a team can spend weeks on demand that was never there.

    This skill keeps one markdown card per customer job. Every entry is a verbatim buyer quote with a source and an evidence level from 0 (a guess, a colleague's opinion, or model output) to 5 (money changed hands). A job exists at 3 or more quotes of level 3 or higher from 2 or more sources; demand exists at 1 or more commitment signal of level 4 or higher. Four modes use the card. Extract pulls quotes from transcripts, reviews, or CRM notes. Translate writes headlines and positioning using only the buyers' own words. Model writes a growth model with a North Star and drivers. Simulate runs a practice interview that never counts as evidence.

    The proof

    The gate is a script, not a judgment. job-card-gate reads a PRD's Job card: and Driver: lines, counts the evidence rows on the card, and checks the driver exists in the growth model. Run against the included fictional example, it prints: job exists PASS, 12 quotes at level 3+ from 5 sources; demand exists PASS, 2 commitment signals at level 4+; driver PASS. The package includes four test fixtures (pass, no card, bad driver, waiver), and the repo's test script checks that each one returns its expected exit code. All quotes in the example and fixtures are synthetic and labeled as such.

    What you get

    SKILL.md with the evidence levels, rules, and four modes; a card template; a fully synthetic worked example for a fictional scheduling product (card, growth model, and PRD); the four gate fixtures; and the gate itself as bin/job-card-gate.py, a Python 3 script with no dependencies.

    How to use

    1. Add the skill folder to your agent's skills directory, and copy the gate onto your PATH as job-card-gate (rename the .py copy or install it from the GitHub repo) and make it executable.
    2. Give the agent a call transcript, review set, or forum thread and ask it to extract the job card.
    3. Add Job card: <path> and Driver: <name> lines to your PRD and run job-card-gate <prd.md> before engineering starts.

    Source and license

    Free to use and MIT licensed. The full source, install steps, and issue tracker are on GitHub at https://github.com/robertmnyborg/analyst-rigor-skills.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Business & Operations.

    The gate only works once job-card-gate is on your PATH; this package ships it as bin/job-card-gate.py because the marketplace requires a file extension, so rename it or install it from the GitHub repo. The gate counts table rows and levels; it cannot tell whether a quote is real, so the verbatim check depends on the extract steps. Extract mode works on text you provide and does not fetch transcripts. The optional auto-nudge hook is not included.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 8 days ago

    • Free with an Agensi account

    Software
    Python 3 for the job-card-gate script

    job-card-gate on your PATH (included as bin/job-card-gate.py). the trustworthy-insights skill, which extract mode calls.

    Permissions

    File Scopes

    project
    /jobs/
    /growth-model.md
    the
    PRD
    or
    spec
    you
    gate

    The gate script only reads files. The agent writes cards, messaging, and growth-model files inside your project folder.

    Listed8 days ago

    What's inside

    Frequently Asked Questions

    An agent can summarize a call into a persona, but it paraphrases, mixes in its own assumptions, and nothing checks the result. This skill keeps only verbatim quotes with a source and level, and the gate script counts them, so a PRD built on simulated or remembered demand fails.

    Yes. The skill is markdown and the gate is a plain Python 3 script, so any agent that reads SKILL.md and can run a shell command works. The optional auto-nudge hook mentioned in SKILL.md is Claude Code-specific and is not included.

    Put a `Job card waiver: <reason, who, date>` line in the PRD. The gate reports WAIVED. The skill will not write a waiver unless you tell it to.

    A model playing your customer returns the answer your team already assumes. Simulate mode turns those answers into interview questions instead.

    On GitHub at https://github.com/robertmnyborg/analyst-rigor-skills, under the MIT license. Every file in this package is there, plus the README and install steps.

    New in Business & Operations

    Free