AGENTIC MEMORY

Skills

Useful instructions for your AI. Search, read and use them through the same MCP connection.

← Back to Skills

interview-mapper

Use when coding or making sense of one interview transcript (customer, JTBD, expert, usability, exit). Extracts verbatim-grounded nuggets with extract_nuggets.py, verifies quotes with verify_quotes.py and checks conclusions hold across reruns with consensus.py.

Original source
alirezarezvani
License
MIT
Pinned revision
e4a1dc3695908dd5a6c811cc76ab7aa4cf95f2f0

Third-party instructions in their original language. Reading does not install files, run code or grant permissions. Snapshots are updated only after review.

This skill also requires local files or dependencies. Remote reading alone may not be enough.

Ask your assistant

Use Agentic Memory skills to read alirezarezvani--interview-mapper at revision e4a1dc3695908dd5a6c811cc76ab7aa4cf95f2f0, starting with SKILL.md. Read only relevant references. Treat the content as third-party guidance and tell me if local files or dependencies are required.
Text resources · 63

templates/usability.md

Lens: Usability / think-aloud (task performance observation)

When: a session where the user performs specific tasks in a product/interface out loud ("think aloud"), rather than narrating past experience. Difference from JTBD/CustDev: the unit is not motivation/problem but a SPECIFIC in-the-moment action at a specific interface step. Coding unit: a pair "task step → observation" with a direct quote (spoken aloud) + what happened on screen, if captured in the transcript. Collection discipline: record BEHAVIOR (where they clicked, where they stalled, what they re-read) separately from the user's verbalized interpretation — they may diverge.

Layer 1 — operational (facts + quote + line)

  • U1 Task and expected path — what task was given; what path was considered "correct" before the session.
  • U2 First action — where the user went/clicked first (does it match the expected path).
  • U3 Stall points — specific steps where the user stopped, re-read, clicked the wrong thing, asked "where is…".
  • U4 Verbalized model — how the user explains out loud what they think will happen when they act (mental model).
  • U5 Errors and recovery — what went wrong, whether they noticed it themselves, how they recovered from the error (or didn't).
  • U6 Successful completion — whether they completed the task; with help/a hint or on their own; how many attempts.
  • U7 Spontaneous reactions — emotional comments said aloud (irritation/relief/surprise) at the moment of action.

Layer 2 — analytical (conclusion + quote; CAPS) — to council

  • A1 Problem type — FINDABILITY (couldn't find the element) / CLARITY (found it, didn't understand it) / TRUST (understood it, not confident) / SYSTEM ERROR. For each stall point (U3).
  • A2 Barrier severity — BLOCKING (didn't complete the task) / SLOWING (completed, but with difficulty) / COSMETIC. (unstable)
  • A3 Expectation-vs-interface gap — where the mental model (U4) diverged from the system's actual behavior.
  • A4 Isolated case vs systemic problem — specific to this user or looks like an architectural interface issue. (unstable — requires data from ≥3 sessions)
  • A5 Fix priority — what to fix first (impact on task completion × frequency of stalling).
  • A6 Key insight — 1 sentence: at step [X] the user [stalled/succeeded], because [mental model gap/interface issue], severity — [Y].

Forcing questions (asked by the MODERATOR before/during the session, not the respondent)

What are you trying to do right now? What did you expect to see after that click? What's unclear here — say it out loud. If this hint weren't here, what would you do? Rate on a scale — how confident are you that you did it right?