interview-debrief

interview-debrief is a command for coding agents from AInsteinsBR/renata. It costs 28 tokens per session (1,308 once invoked), scanned A, original, MIT.

A transcript-review command for problem interviews, which are conversations used to learn whether people have a real problem. It extracts exact quotes, grades the evidence, updates the evidence board, and gives feedback on the interviewer's questions using Mom Test rules.

In plain words
What is it for?
Use it with pasted text or a text/Markdown transcript to classify signals, update evidence records, promote or demote evidence seals, and improve interview technique.
Why use it?
It turns a conversation into traceable evidence instead of relying on memory or impressions. It also helps prevent leading questions and other interview mistakes in future conversations.

Command

Part of the renata plugin — 3 skills, 33 commands, 6 agents, 2 hooks shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/ainsteinsbr/renata/interview-debrief
Clone the repo
git clone --depth 1 https://github.com/AInsteinsBR/renata

Or install renata, the plugin that ships this one along with the rest of its 3 skills, 33 commands, 6 agents, 2 hooks.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for interview-debrief

README.md
[![agentmods](https://agentmods.dev/badge/commands/ainsteinsbr/renata/interview-debrief.svg)](https://agentmods.dev/commands/ainsteinsbr/renata/interview-debrief)
Your own site
<a href="https://agentmods.dev/commands/ainsteinsbr/renata/interview-debrief"><img src="https://agentmods.dev/badge/commands/ainsteinsbr/renata/interview-debrief.svg" alt="Measured on agentmods" height="20"></a>
Per session 28 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,308 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00028 $0.01308
Opus 5 $0.00014 $0.00654
Sonnet 5 $0.00006 $0.00262
Haiku 4.5 $0.00003 $0.00131

Measured 4d ago against content hash be4e1a0f74ce, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

interview-debrief scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

commands/interview-debrief.md · 104 lines

How it starts

The opening of the file, as written. The whole thing — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/renata:interview-debrief — From transcript to evidence (and better questions next time)

You are an honest, skeptical researcher (Mom Test school). You take a ready transcript of a problem interview and turn it into evidence: verbatim quotes classified by strength, an updated evidence board, promoted (or demoted) evidence seals — plus mandatory coaching on the interviewer's own questions, so the next interview is better than this one.

Respond to the user and generate document content in the user's language (the language they are writing in).

It is the after half of the interview loop (/renata:interview-kit is the before). The verdict on the assumption itself belongs to /renata:assumption-test — this command only accumulates and grades the evidence.

Input rule (hard)

  • Ready transcript ONLY: pasted text or a path to a text/markdown file.
  • If given an audio/video file path (.mp3, .m4a, .wav, .mp4, .ogg): politely refuse — "Give me the transcript — your recorder/Meet/Zoom already produces one." Raw-audio integration is a declared future evolution, not a v1 gap to work around.

Before generating

  1. Get the transcript ($ARGUMENTS = text or file path; if missing, ask for it).
  2. Read docs/interviews/kits/*.md — find the matching kit (ask if ambiguous). No kit is acceptable (interview happened before the kit existed) — then ask which assumption(s) it targets.
  3. Read the source docs the kit points to (docs/assumptions/*.md / docs/discovery/*.md).
  4. Ask ONE at a time (only what is missing): who was interviewed (profile/role, not name — keep transcripts pseudonymous) and when.

Signal classification (the core)

Every signal is a verbatim quote — paraphrase is NOT evidence. Classify each:

Class Meaning Counts as evidence?
🥇 spontaneous the person brought it up unprompted Yes — strongest
🥈 prompted direct answer to a question Weak — supporting only
🚫 contaminated came after a pitch / leading question No — discard, and flag it in coaching

Read the full file on GitHub · 104 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 104 lines · 28 tokens per session scan A be4e1a0f74ce

Subscribe to this mod's changes

interview-debrief is a command published in the GitHub repository AInsteinsBR/renata (10 stars, last pushed 13d ago), licensed MIT. It adds 28 tokens to every session and 1,308 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.