# Chapter 2 lab — captured receipts and the command card

This chapter’s turns were executed for real (Linux, Ollama 0.40.8 serving gemma4:e4b, pytest 9.1.1, Go-built relurpish from the tree that became BETA-RC1-FALL2026). The session was driven through a correctly sized pseudo-terminal by an expect-style script (drive_demo.py in the author’s capture kit), keystroke by keystroke: boot → task → approval banner → s → completion. Everything the chapter quotes exists on disk in the demo workspace.

# The two turns, as captured

# Reading your own receipts

Run from the stringkit workspace after a turn:

# why did that route run? (decided_by, family, candidate_count)
jq -c 'select(.resourceLogs) | .resourceLogs[]?.scopeLogs[]?.logRecords[]?
       | select(.eventName=="euclo.route.selected")' \
    .relurpify_state/telemetry/otlp/*.jsonl | tail -1

# how did the turn end — really? (outcome, llm_calls)
jq -c 'select(.resourceLogs) | .resourceLogs[]?.scopeLogs[]?.logRecords[]?
       | select(.eventName=="euclo.execution.complete")' \
    .relurpify_state/telemetry/otlp/*.jsonl | tail -1

# the approval record lives in the chained audit log
ls .relurpify_state/audit/

# the fact-check
python3 -m pytest test_stringkit.py
git diff

Known documentation bug, reproduced: the shorter filter jq '.. | .logRecords[]?' (seen in some project docs) fails on jq 1.8 with Cannot index array with string. Use the select(.resourceLogs) form above.

# What this lab proves about the product

• The documented demo replicates end to end: gate opens, turn classifies, selects, pauses for approval, writes governed, bounds, completes — all inside a minute on the captured hardware.
• outcome: "success" records a completed execution; a false completion claim passed through it while pytest showed SyntaxError. The runtime’s receipt and the repository’s health are different documents. Chapter 2 teaches the difference; Chapter 9 pins it with deterministic tests.