Reads seven constitutional files (TELOS, DA_IDENTITY, PRINCIPAL_IDENTITY, PROJECTS, system prompt, PRINCIPAL_TELOS, ARCHITECTURE_SUMMARY) via TelosFreshness, surfaces stalest items, drives contextual peer conversation.
A fresh AI session doesn't know your goals, your projects, or how your thinking has shifted since the last time you talked. You can paste in context manually every time, but it goes stale, and the model has no way to tell which parts are still true. After a few weeks, your TELOS file drifts from reality — sections you've moved past stay current, constraints you've resolved never get updated, and the DA keeps optimizing for a version of you from three months ago.
Interview reads seven constitutional files (TELOS, DA_IDENTITY, PRINCIPAL_IDENTITY, PROJECTS, system prompt, PRINCIPAL_TELOS, ARCHITECTURE_SUMMARY) and checks them against staleness thresholds via TelosFreshness. A routing decision runs first: if the scan finds Phase 0 targets with completeness below 80 (DA name still 'PAI', identity still 'User', projects still sample rows), it runs Phase0Setup to get the system grounded before doing anything else. For populated installs, it runs ContextCheckin — surfacing the stalest sections by age and driving a focused peer conversation: one entry at a time, per typed ID on structured sections (G3, M0, P2), section-level on prose. Stop signals ('enough', 'later') exit cleanly; every approved edit bumps the freshness timestamp via TelosFreshness.ts so the staleness signal stays accurate. ContextAudit surfaces TBD markers left in files. The whole thing is a context debt payment, not a form fill.
What you say to your DA, and what the Interview skill actually does.
The thinking, frameworks, and architecture that distinguish this skill from a generic version of the same task.
Interview reads your constitutional files — TELOS, identity, projects, system prompt, architecture — checks their freshness, surfaces the stalest items, and drives a contextual peer conversation to refresh them. On a populated system it runs a context check-in; on a fresh install it falls back to first-time setup.
The files that define who you are and what you're working toward drift out of date the moment you stop looking at them. A goal you set in January may be done, dead, or still right — but nobody re-reads the whole TELOS to find out, so the context the system runs on slowly rots. The usual fix is a blank "what's your mission?" prompt, which ignores everything already on file and makes you repeat yourself. This skill reads what's there first, flags only what's gone stale, and asks "still right?" instead of starting from zero.
The skill reads the constitutional files via the freshness tooling, scores each section's staleness, and routes to the right workflow — a check-in on a populated system, or first-time setup on a fresh one. Staleness is a priority signal, not a failure flag.
bun ~/.claude/PAI/TOOLS/TelosFreshness.ts --bump <slug>. Without this the staleness signal degrades to noise.last_updated:) returns fileUpdated: null and every section reads as stale. Run bun ~/.claude/PAI/TOOLS/MigrateTelosFreshness.ts once; idempotent and content-preserving (verifies sha256 of stripped content).current_state, "Wrong (Things I've been wrong about)" → wrong, "2036 — A Day in the Life…" → 2036. Always run heading text through sectionSlug() from TelosFreshness.ts./api/telos/freshness/summary call returns the cached value until invalidation. Send /reload (POST) to invalidate the cache immediately, or wait 60s.GenerateTelosSummary.ts) preserves the markers. It splits by ^## headings; the per-section HTML comments live inside the section body and are not re-emitted in PRINCIPAL_TELOS.md. Safe to run after edits.InterviewScan.ts calls readTelosFreshness() once at startup and adds age_days, threshold_days, stale to every Phase 1+ target row. Stale sections get a +200 priority bump so they naturally rise in --next output./Telos — edit a single TELOS section directly (without the conversational walk)./Migrate — intake content from other sources (one-shot classification, not an interview)./_PROFILE — manage PRINCIPAL_IDENTITY directly.Skill("ISA") — interview an ISA (different artifact, different workflow).Workflows/ContextCheckin.md default /interview on a populated system; context check-in, telos check-in, whats stale, how are we doing on…, still on…, review context
Workflows/Phase0Setup.md fresh install; DA name still reads PAI; PRINCIPAL_IDENTITY still reads User; PROJECTS sample-row only; .env missing required keys
Workflows/TelosCheckin.md` (redirects to ContextCheckin) back-compat for explicit telos checkin routing
Say any of these to your DA and PAI activates the Interview skill automatically:
Or invoke explicitly:
Skill("Interview")Install PAI on your machine — your DA gets the Interview skill plus 44 others, all hooked into one Life OS.