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/team-retro.md

Lens: Team retrospective (post-mortem)

When: group discussion of a COMPLETED project/sprint with the team that worked on it. Distinction from a focus group: the topic isn't opinions about a product/brand, but shared experience of the work process. Distinction from ВМДИ org-mapping: not about the system as a whole, but about one specific finished project. Coding unit: utterance + speaker + which project phase it relates to (planning/execution/delivery). Collection discipline: look for concrete episodes ("at moment X, Y happened"), not general assessments ("overall it went fine"). Team disagreement about what went wrong isn't a protocol bug — it's a signal.

Layer 1 — operational (facts + quote + line + speaker)

  • T1 Project context — what they did, timeline, team composition, roles of retro participants.
  • T2 What worked (keep) — concrete practices/decisions the team wants to keep; who proposed them.
  • T3 What didn't work (stop) — concrete episodes of process failure, with the phase noted (planning/execution/delivery) and who raised it.
  • T4 What to try (try) — new ideas for next time; whose suggestions, whether the group supported them.
  • T5 Key decisions and forks — moments where the team chose between options; who insisted, whether the choice was vindicated in hindsight.
  • T6 Cross-role friction — where role interests/priorities diverged (e.g. design vs development vs product), concrete episode.
  • T7 External blockers — what got in the way from outside (other teams, client, resources) — outside the team's control.

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

  • A1 Failure-cause attribution — PROCESS / COMMUNICATION / RESOURCES / EXTERNAL / PERSONAL. Per each T3. (unstable)
  • A2 Team agreement on cause — do different participants' explanations of the same episode match, or does each see a different cause (divergence is itself a finding).
  • A3 Reproducibility of success (T2) — a systemic practice repeatable on any project, vs a one-off due to this team's makeup.
  • A4 Unresolved conflict — friction (T6) the team voiced but didn't resolve — risk of recurrence on the next project.
  • A5 Change priority — which item from T4 to implement first by effect/effort ratio, judging by unanimity of support.
  • A6 Key insight — 1 sentence: project [name] — key bottleneck cause [X] (attribution [Y]), team agrees/disagrees on the explanation, next step — [Z].

Forcing questions (for the FACILITATOR)

What would you do differently if you started over today? At what point did it become clear something was going wrong? Who else saw this problem at the time — why wasn't it raised sooner? What do you disagree with in what your colleagues just said?

Distinction from S3 (reliability council)

Divergence of participants' views on the same episode (A2) is data about the team — record it as is, don't smooth it into one "consensus" conclusion. The S3 council (independent AI runs) is needed separately for unstable analytical cells (A1, A3); it doesn't substitute for recording real team disagreement.