lab-kit-cli 0.4.0__tar.gz → 0.6.0__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (207) hide show
  1. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/CHANGELOG.md +15 -0
  2. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/PKG-INFO +3 -3
  3. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/README.md +1 -1
  4. {lab_kit_cli-0.4.0/examples/monte-carlo-lab/.agents → lab_kit_cli-0.6.0}/agents/runner.md +1 -1
  5. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/docs/spec/lab-model.md +34 -34
  6. lab_kit_cli-0.6.0/examples/monte-carlo-lab/.agents/agents/runner.md +41 -0
  7. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/.agents/skills/experiment/SKILL.md +4 -4
  8. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0/examples/monte-carlo-lab/.agents}/skills/plan-mission/SKILL.md +1 -1
  9. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0/examples/monte-carlo-lab/.agents}/skills/review/SKILL.md +2 -2
  10. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0/examples/monte-carlo-lab/.agents}/skills/set-up-lab/SKILL.md +1 -1
  11. lab_kit_cli-0.6.0/examples/monte-carlo-lab/.gitignore +4 -0
  12. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/.lab/method/DISCIPLINE.md +1 -1
  13. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0/examples/monte-carlo-lab/.lab}/method/LADDER.md +2 -2
  14. lab_kit_cli-0.6.0/examples/monte-carlo-lab/.lab/migration.tsv +4 -0
  15. lab_kit_cli-0.6.0/examples/monte-carlo-lab/docs/.folio/catalog.json +491 -0
  16. lab_kit_cli-0.6.0/examples/monte-carlo-lab/docs/.folio/search.json +248 -0
  17. lab_kit_cli-0.6.0/examples/monte-carlo-lab/docs/baseline.sha256 +9 -0
  18. lab_kit_cli-0.6.0/examples/monte-carlo-lab/docs/content/journal/2026/2026-10-04-concluded-the-pi-error-scaling-mission.md +10 -0
  19. lab_kit_cli-0.6.0/examples/monte-carlo-lab/docs/content/journal/2026/2026-10-04-locked-the-pi-error-scaling-protocol.md +10 -0
  20. lab_kit_cli-0.6.0/examples/monte-carlo-lab/docs/content/journal/2026/2026-10-04-opened-the-pi-error-scaling-mission.md +10 -0
  21. lab_kit_cli-0.6.0/examples/monte-carlo-lab/docs/content/journal/2026/2026-10-04-ran-the-pi-error-scaling-protocol.md +10 -0
  22. lab_kit_cli-0.6.0/examples/monte-carlo-lab/docs/content/journal/2026/2026-10-04-recorded-r-1.md +9 -0
  23. lab_kit_cli-0.6.0/examples/monte-carlo-lab/docs/content/results/R-1.md +18 -0
  24. lab_kit_cli-0.6.0/examples/monte-carlo-lab/docs/content/results/R-2.md +19 -0
  25. lab_kit_cli-0.6.0/examples/monte-carlo-lab/docs/folio.yaml +9 -0
  26. lab_kit_cli-0.6.0/examples/monte-carlo-lab/experiments/26-10-04-e_pi-error-scaling/lock.json +11 -0
  27. lab_kit_cli-0.6.0/examples/monte-carlo-lab/experiments/26-10-04-e_pi-error-scaling/score.yaml +50 -0
  28. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/lab.yaml +1 -2
  29. lab_kit_cli-0.6.0/examples/monte-carlo-lab/missions/26-10-04-m_pi-error-scaling.md +42 -0
  30. lab_kit_cli-0.6.0/examples/monte-carlo-lab/runs/26-10-04-e_pi-error-scaling/26-10-04-2316-r_zlk7/run.json +43 -0
  31. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/method/DISCIPLINE.md +1 -1
  32. lab_kit_cli-0.6.0/method/LADDER.md +73 -0
  33. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/pyproject.toml +2 -2
  34. lab_kit_cli-0.6.0/skills/experiment/SKILL.md +80 -0
  35. lab_kit_cli-0.6.0/skills/plan-mission/SKILL.md +79 -0
  36. lab_kit_cli-0.6.0/skills/review/SKILL.md +75 -0
  37. lab_kit_cli-0.6.0/skills/set-up-lab/SKILL.md +81 -0
  38. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/__init__.py +1 -1
  39. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/checks/files.py +25 -103
  40. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/checks/locks.py +43 -41
  41. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/checks/results.py +26 -10
  42. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/checks/run.py +1 -2
  43. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/cli.py +30 -18
  44. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/commands/experiments.py +22 -16
  45. lab_kit_cli-0.6.0/src/lab_kit/commands/migrate.py +661 -0
  46. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/commands/runs.py +31 -22
  47. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/commands/setup.py +27 -3
  48. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/lab.py +38 -4
  49. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/ops.py +12 -2
  50. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/records.py +111 -40
  51. lab_kit_cli-0.6.0/src/lab_kit/store.py +273 -0
  52. lab_kit_cli-0.6.0/src/lab_kit/supervise.py +132 -0
  53. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/.agents/agents/reporter.md +30 -0
  54. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/.agents/agents/reviewer.md +36 -0
  55. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/.agents/agents/scout.md +28 -0
  56. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/.agents/skills/address/SKILL.md +75 -0
  57. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/.agents/skills/configure/SKILL.md +88 -0
  58. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/.agents}/skills/experiment/SKILL.md +3 -3
  59. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/.agents/skills/organise/SKILL.md +92 -0
  60. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/.agents/skills/publish/SKILL.md +70 -0
  61. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/.agents/skills/review/SKILL.md +2 -2
  62. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/.agents/skills/run/SKILL.md +69 -0
  63. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/.agents/skills/run-mission/SKILL.md +83 -0
  64. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/.agents/skills/set-up/SKILL.md +72 -0
  65. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/.agents/skills/write/SKILL.md +111 -0
  66. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/.lab/frozen.sha256 +2 -0
  67. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/.lab/method/DISCIPLINE.md +114 -0
  68. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/AGENTS.md +22 -0
  69. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/CLAUDE.md +1 -0
  70. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/.folio/backlinks.json +425 -0
  71. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/.folio/cards.json +1 -0
  72. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/.folio/journal.json +143 -0
  73. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/.folio/nav.json +1 -0
  74. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/.gitignore +4 -0
  75. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/assets/figures/.gitkeep +0 -0
  76. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/assets/refs.bib +0 -0
  77. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/content/claims/C-1.md +26 -0
  78. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/content/index.html +19 -0
  79. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/content/journal/2026/2026-10-04-lab-set-up.md +10 -0
  80. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/content/journal/2026/2026-10-04-r-1-fitted-the-mean-absolute-error.md +10 -0
  81. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/content/journal/2026/2026-10-04-recorded-r-2.md +9 -0
  82. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/content/journal/2026/2026-10-04-reported-the-pi-error-scaling-experiment.md +9 -0
  83. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/content/journal/2026/2026-10-04-the-square-root-law-holds-per-size-predictions.md +10 -0
  84. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/content/maps/research.html +29 -0
  85. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/content/projects/lab/index.html +31 -0
  86. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/content/protocols/pi-error-scaling.md +66 -0
  87. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/content/questions/Q-1.md +25 -0
  88. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/content/reports/pi-error-scaling-report/index.html +47 -0
  89. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/experiments/.gitkeep +0 -0
  90. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/experiments/pi-error-scaling/bin/.gitkeep +0 -0
  91. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/experiments/pi-error-scaling/bin/fold.py +101 -0
  92. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/experiments/pi-error-scaling/bin/sample.py +65 -0
  93. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/experiments/pi-error-scaling/runs/.gitkeep +0 -0
  94. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/experiments/pi-error-scaling/runs/20261004-231629/out/estimates.tsv +501 -0
  95. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/lab.yaml +5 -0
  96. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/ops/STATE.md +7 -0
  97. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/ops/missions/.gitkeep +0 -0
  98. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/substrate/README.md +3 -0
  99. lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/substrate/estimator.py +18 -0
  100. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/tests/test_budgets.py +2 -2
  101. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/tests/test_checks.py +23 -54
  102. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/tests/test_end_to_end.py +12 -16
  103. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/tests/test_example.py +1 -1
  104. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/tests/test_instrument.py +4 -5
  105. lab_kit_cli-0.6.0/tests/test_migrate.py +206 -0
  106. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/tests/test_runs.py +4 -4
  107. lab_kit_cli-0.6.0/tests/test_store.py +115 -0
  108. lab_kit_cli-0.4.0/src/lab_kit/store.py +0 -285
  109. lab_kit_cli-0.4.0/src/lab_kit/supervise.py +0 -103
  110. lab_kit_cli-0.4.0/tests/test_store.py +0 -108
  111. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/.github/workflows/ci.yml +0 -0
  112. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/.github/workflows/policy.yml +0 -0
  113. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/.github/workflows/release.yml +0 -0
  114. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/.gitignore +0 -0
  115. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/AGENTS.md +0 -0
  116. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/CLAUDE.md +0 -0
  117. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/CONTRIBUTING.md +0 -0
  118. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/LICENSE +0 -0
  119. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/SETUP.md +0 -0
  120. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/agents/reporter.md +0 -0
  121. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/agents/reviewer.md +0 -0
  122. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/agents/scout.md +0 -0
  123. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/docs/figures/how-lab-kit-works-dark.svg +0 -0
  124. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/docs/figures/how-lab-kit-works-light.svg +0 -0
  125. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/docs/figures/lab-loop-dark.svg +0 -0
  126. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/docs/figures/lab-loop-light.svg +0 -0
  127. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/docs/figures/number-chain-dark.svg +0 -0
  128. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/docs/figures/number-chain-light.svg +0 -0
  129. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/.agents/agents/reporter.md +0 -0
  130. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/.agents/agents/reviewer.md +0 -0
  131. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/.agents/agents/scout.md +0 -0
  132. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/.agents/skills/address/SKILL.md +0 -0
  133. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/.agents/skills/configure/SKILL.md +0 -0
  134. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/.agents/skills/organise/SKILL.md +0 -0
  135. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/.agents/skills/publish/SKILL.md +0 -0
  136. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/.agents/skills/run/SKILL.md +0 -0
  137. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/.agents/skills/run-mission/SKILL.md +0 -0
  138. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/.agents/skills/set-up/SKILL.md +0 -0
  139. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/.agents/skills/write/SKILL.md +0 -0
  140. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/.lab/frozen.sha256 +0 -0
  141. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/AGENTS.md +0 -0
  142. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/CLAUDE.md +0 -0
  143. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/.folio/backlinks.json +0 -0
  144. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/.folio/cards.json +0 -0
  145. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/.folio/journal.json +0 -0
  146. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/.folio/nav.json +0 -0
  147. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/.gitignore +0 -0
  148. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/assets/data/.gitkeep +0 -0
  149. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/assets/figures/.gitkeep +0 -0
  150. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/assets/refs.bib +0 -0
  151. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/content/claims/C-1.md +0 -0
  152. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/content/index.html +0 -0
  153. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/content/journal/2026/2026-10-04-lab-set-up.md +0 -0
  154. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/content/journal/2026/2026-10-04-r-1-fitted-the-mean-absolute-error.md +0 -0
  155. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/content/journal/2026/2026-10-04-recorded-r-2.md +0 -0
  156. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/content/journal/2026/2026-10-04-reported-the-pi-error-scaling-experiment.md +0 -0
  157. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/content/journal/2026/2026-10-04-the-square-root-law-holds-per-size-predictions.md +0 -0
  158. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/content/maps/research.html +0 -0
  159. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/content/projects/lab/index.html +0 -0
  160. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/content/protocols/pi-error-scaling.md +0 -0
  161. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/content/questions/Q-1.md +0 -0
  162. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/docs/content/reports/pi-error-scaling-report/index.html +0 -0
  163. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/experiments/.gitkeep +0 -0
  164. {lab_kit_cli-0.4.0/examples/monte-carlo-lab/experiments/pi-error-scaling → lab_kit_cli-0.6.0/examples/monte-carlo-lab/experiments/26-10-04-e_pi-error-scaling}/bin/.gitkeep +0 -0
  165. {lab_kit_cli-0.4.0/examples/monte-carlo-lab/experiments/pi-error-scaling → lab_kit_cli-0.6.0/examples/monte-carlo-lab/experiments/26-10-04-e_pi-error-scaling}/bin/fold.py +0 -0
  166. {lab_kit_cli-0.4.0/examples/monte-carlo-lab/experiments/pi-error-scaling → lab_kit_cli-0.6.0/examples/monte-carlo-lab/experiments/26-10-04-e_pi-error-scaling}/bin/sample.py +0 -0
  167. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/ops/STATE.md +0 -0
  168. {lab_kit_cli-0.4.0/examples/monte-carlo-lab/experiments/pi-error-scaling/runs → lab_kit_cli-0.6.0/examples/monte-carlo-lab/ops/missions}/.gitkeep +0 -0
  169. {lab_kit_cli-0.4.0/examples/monte-carlo-lab/experiments/pi-error-scaling/runs/20261004-231629 → lab_kit_cli-0.6.0/examples/monte-carlo-lab/runs/26-10-04-e_pi-error-scaling/26-10-04-2316-r_zlk7}/out/estimates.tsv +0 -0
  170. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/substrate/README.md +0 -0
  171. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/examples/monte-carlo-lab/substrate/estimator.py +0 -0
  172. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/skills/run-mission/SKILL.md +0 -0
  173. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/__main__.py +0 -0
  174. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/budgets.py +0 -0
  175. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/checks/__init__.py +0 -0
  176. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/checks/base.py +0 -0
  177. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/commands/__init__.py +0 -0
  178. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/commands/status.py +0 -0
  179. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/data.py +0 -0
  180. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/errors.py +0 -0
  181. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/frozen.py +0 -0
  182. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/gitlog.py +0 -0
  183. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/protocol.py +0 -0
  184. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/src/lab_kit/rederive.py +0 -0
  185. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0}/tests/conftest.py +0 -0
  186. {lab_kit_cli-0.4.0 → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/.agents}/agents/runner.md +0 -0
  187. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/.agents/skills/plan-mission/SKILL.md +0 -0
  188. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/.agents/skills/set-up-lab/SKILL.md +0 -0
  189. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/.gitignore +0 -0
  190. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/.lab/method/LADDER.md +0 -0
  191. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/docs/.folio/catalog.json +0 -0
  192. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/docs/.folio/search.json +0 -0
  193. {lab_kit_cli-0.4.0/examples/monte-carlo-lab/ops/missions → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5/docs/assets/data}/.gitkeep +0 -0
  194. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/docs/content/journal/2026/2026-10-04-concluded-the-pi-error-scaling-mission.md +0 -0
  195. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/docs/content/journal/2026/2026-10-04-locked-the-pi-error-scaling-protocol.md +0 -0
  196. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/docs/content/journal/2026/2026-10-04-opened-the-pi-error-scaling-mission.md +0 -0
  197. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/docs/content/journal/2026/2026-10-04-ran-the-pi-error-scaling-protocol.md +0 -0
  198. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/docs/content/journal/2026/2026-10-04-recorded-r-1.md +0 -0
  199. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/docs/content/results/R-1.md +0 -0
  200. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/docs/content/results/R-2.md +0 -0
  201. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/docs/folio.yaml +0 -0
  202. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/experiments/pi-error-scaling/.gitignore +0 -0
  203. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/experiments/pi-error-scaling/lock.json +0 -0
  204. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/experiments/pi-error-scaling/runs/20261004-231629/MANIFEST.sha256 +0 -0
  205. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/experiments/pi-error-scaling/runs/20261004-231629/run.json +0 -0
  206. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/experiments/pi-error-scaling/score.yaml +0 -0
  207. {lab_kit_cli-0.4.0/examples/monte-carlo-lab → lab_kit_cli-0.6.0/tests/fixtures/monte-carlo-lab-0.5}/ops/missions/2026-10-04-pi-error-scaling.md +0 -0
@@ -1,5 +1,20 @@
1
1
  # Changelog
2
2
 
3
+ ## 0.6.0
4
+
5
+ - One folder per run, outside git. A run is `runs/<experiment>/<YY-MM-DD-HHMM>-r_<4 characters>/` and holds everything it produced. lab-kit writes only `run.json` and `run.log` there; the rest is the experiment's own. `LAB_RUN` names the folder. `LAB_OUT`, `LAB_WORK` and `LAB_RUN_DIR` still work for one release.
6
+ - The seal. When the command exits, `run.json` records every file's sha256 in `files`, and one hash over the whole record in `seal`. `MANIFEST.sha256`, `work.sha256` and `store.json` are gone. `lab-evidence-sealed` checks a run's folder and its record against the seal.
7
+ - The store holds the runs. With `store:`, `runs/` is a committed link to `<store>/runs`. A run is written on local disk in `.lab/running/`, then moves into the store when it ends: copied, verified against its seal, renamed into place, made read-only, and only then deleted locally. `lab-kit publish --pending` retries a failed move, `lab-kit fetch` copies a run into `.lab/fetched/` on a machine without the store, and `lab-kit verify` checks a run against its seal. Where `runs/` does not resolve, as in CI, the gate skips the checks that read runs and says so once. `lab-work-stored` becomes `lab-run-stored`.
8
+ - Dated ids. Missions are `missions/<YY-MM-DD>-m_<slug>.md` (they were in `ops/missions/`), and experiments are `experiments/<YY-MM-DD>-e_<slug>/`. A protocol keeps its slug.
9
+ - `lab-kit migrate [--apply]` moves a lab from the 0.5.0 layout. It takes each experiment's date from its lock and each run's from its start. Each run's `out/`, `work/` and `run.log` go into one sealed folder in `runs/`. Every old path and id in git-tracked text is rewritten, and the lock hashes are taken again where a protocol's text changed with it. The rewritten documents are listed in the library's `baseline.sha256`, folio 0.3.6's `baseline`, so their permanence holds from the rewrite on, and every old name is kept in `.lab/migration.tsv`. Run it again after old runs land from machines on an older lab-kit.
10
+ - Run records are no longer append-only lab files in git: the seal holds them.
11
+
12
+ ## 0.5.0
13
+
14
+ - lab-kit no longer runs a lab's code to check it. The `lab-selftest` check is gone, and with it the `tools:` key in `lab.yaml`: a lab that still has the key gets an error saying to delete it. lab-kit checks the record; the lab tests its code in its own test suite, which its gate runs beside `lab-kit check`.
15
+ - The method (DISCIPLINE §7) and the experiment and review skills say so: an experiment's command calls the lab's own tested code, `bin/` holds only what is particular to one experiment and may be empty, and a fold is tested on a planted case in the lab's suite. The lock still hashes `bin/` and records the commit, which says which version of the lab's code a run used.
16
+ - Locks and run records are unchanged.
17
+
3
18
  ## 0.4.0
4
19
 
5
20
  - The run store. `store:` in `lab.yaml` names one shared filesystem, such as `/labs/<lab>`, where every finished run's raw output goes. When a run ends, `lab-kit run` publishes `work/` and `run.log` to the store at the run folder's own path: a copy into `.incoming/`, checked against `work.sha256` (nothing missing, extra or changed), renamed into place, made read-only. It then writes `store.json` beside `run.json` and deletes the local copy. A run in the store is never overwritten. A failed publish is noted in `run.log`, keeps the local copy and never changes the run's exit code.
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.5
2
2
  Name: lab-kit-cli
3
- Version: 0.4.0
3
+ Version: 0.6.0
4
4
  Summary: A research lab's method and machinery on top of folio: pre-register, lock, run, score, re-derive.
5
5
  Project-URL: Homepage, https://github.com/pierg/lab-kit
6
6
  Project-URL: Issues, https://github.com/pierg/lab-kit/issues
@@ -8,7 +8,7 @@ Author: Piergiuseppe Mallozzi
8
8
  License-Expression: MIT
9
9
  License-File: LICENSE
10
10
  Requires-Python: >=3.10
11
- Requires-Dist: folio-kb>=0.1.0
11
+ Requires-Dist: folio-kb>=0.3.6
12
12
  Requires-Dist: pyyaml
13
13
  Provides-Extra: dev
14
14
  Requires-Dist: pytest; extra == 'dev'
@@ -82,7 +82,7 @@ The gate, `lab-kit check`, runs folio's checks and then the lab's. It runs offli
82
82
  - **Runs.** Every run started after its lock, used exactly the pinned configuration, and its evidence matches its manifest (`lab-run-after-lock`, `lab-roster-frozen`, `lab-evidence-sealed`).
83
83
  - **Results.** Every live result re-derives exactly, rests on a finished run of a locked protocol, and passed an independent reviewer (`lab-rederive`, `lab-result-grounded`).
84
84
  - **Scores.** A scorecard scores exactly the ids the lock names (`lab-score-exact`), and a scored experiment is reported or explained (`lab-scored-reported`).
85
- - **Lab files.** Ids resolve, frozen surfaces are untouched, scripts pass `--selftest`, nothing runs under an unapproved mission, spend stays within its cap and its standing budget, records only grow, the method is current, and the state file stays a short pointer.
85
+ - **Lab files.** Ids resolve, frozen surfaces are untouched, nothing runs under an unapproved mission, spend stays within its cap and its standing budget, records only grow, the method is current, and the state file stays a short pointer.
86
86
 
87
87
  Together they hold a chain from every number on a page back to the lock:
88
88
 
@@ -66,7 +66,7 @@ The gate, `lab-kit check`, runs folio's checks and then the lab's. It runs offli
66
66
  - **Runs.** Every run started after its lock, used exactly the pinned configuration, and its evidence matches its manifest (`lab-run-after-lock`, `lab-roster-frozen`, `lab-evidence-sealed`).
67
67
  - **Results.** Every live result re-derives exactly, rests on a finished run of a locked protocol, and passed an independent reviewer (`lab-rederive`, `lab-result-grounded`).
68
68
  - **Scores.** A scorecard scores exactly the ids the lock names (`lab-score-exact`), and a scored experiment is reported or explained (`lab-scored-reported`).
69
- - **Lab files.** Ids resolve, frozen surfaces are untouched, scripts pass `--selftest`, nothing runs under an unapproved mission, spend stays within its cap and its standing budget, records only grow, the method is current, and the state file stays a short pointer.
69
+ - **Lab files.** Ids resolve, frozen surfaces are untouched, nothing runs under an unapproved mission, spend stays within its cap and its standing budget, records only grow, the method is current, and the state file stays a short pointer.
70
70
 
71
71
  Together they hold a chain from every number on a page back to the lock:
72
72
 
@@ -17,7 +17,7 @@ Execution:
17
17
  - Launch every run with `lab-kit run`. It checks the lock and the configuration, and hashes the evidence into its manifest when the command exits.
18
18
  - Copy the mechanics of the earlier experiment the protocol names: its scripts, its pins, its gitignore. Do not invent new ones.
19
19
  - Verify every recovered artefact against its expected hash. Report a mismatch; never substitute silently.
20
- - Write only in the experiment's `bin/` and in a run's `work/`. Evidence reaches `out/` through the run's own command.
20
+ - Write only in the experiment's `bin/` and in the run's own folder (`$LAB_RUN`). Evidence reaches it through the run's own command.
21
21
  - Fail loud. An apparatus failure is a result to report, not a thing to patch around.
22
22
 
23
23
  Reporting:
@@ -43,21 +43,20 @@ A lab is a repository. Its folio library sits at the root or in a folder such as
43
43
  reports/{slug}/ one experiment's outcome in plain English; frozen once live
44
44
  journal/{yyyy}/ the library's one journal: one permanent file per entry, {date}-{slug}.md
45
45
  ... maps, concepts, the front door, papers
46
- experiments/{slug}/ one per protocol, same slug
47
- bin/ the lab's own scripts for this experiment, each with --selftest
46
+ missions/{YY-MM-DD}-m_{slug}.md one per operator objective; its log is append-only
47
+ experiments/{YY-MM-DD}-e_{slug}/ one per protocol: the day it was created, then the protocol's slug
48
+ bin/ what is particular to this experiment, if anything; may be empty
48
49
  lock.json the lock record: protocol id, its sha256 with the status field set aside, the lock time (UTC),
49
50
  the instrument (a hash of bin/), the git commit and the lab-kit version
50
- runs/{run-id}/
51
- run.json protocol hash, frozen configuration, mission, start, end, exit code, commit, lab-kit version
52
- out/ the evidence, committed at the end of the run, never edited
53
- MANIFEST.sha256 a hash of every file in out/, written when the run ends
54
- work/ run.log raw output and log, gitignored
55
- work.sha256 a hash of every file in work/, written when the run ends; committed
56
- store.json where work/ and run.log were published, and when (when lab.yaml names a store); committed
57
51
  score.yaml the review's scorecard against the lock, and the results to record
52
+ runs/ every run, never in git: a link to <store>/runs when lab.yaml names a store
53
+ {experiment}/{YY-MM-DD-HHMM}-r_{4 characters}/ one run: its start (UTC) and four random letters or digits
54
+ run.json written by lab-kit: experiment, protocol, lock hash, configuration, mission, command, host,
55
+ start, end, exit code, commit, lab-kit version, and the seal: every file's hash
56
+ run.log written by lab-kit: the command's output
57
+ ... whatever the run's command writes; its layout is the experiment's own
58
58
  ops/
59
59
  STATE.md what is true now and what is next; a pointer, never a record
60
- missions/{date}-{slug}.md one per operator objective; its log is append-only
61
60
  budgets/{name}.yaml a standing budget: the operator's grants for many missions; grants only grow
62
61
  .lab/
63
62
  method/ DISCIPLINE.md and LADDER.md, copied by lab-kit, never hand-edited
@@ -70,27 +69,28 @@ A lab is a repository. Its folio library sits at the root or in a folder such as
70
69
  `lab.yaml`:
71
70
 
72
71
  ```yaml
73
- lab-kit: 0.4.0 # the version this lab is checked with
72
+ lab-kit: 0.6.0 # the version this lab is checked with
74
73
  library: docs # where folio.yaml is; "." for the root
75
74
  front: lab # the front door's id: a project document
76
75
  frozen: # surfaces never edited in place
77
76
  - substrate/
78
- tools: [tools/fold.py] # lab-wide scripts; each must pass --selftest
79
77
  rederive_timeout: 120 # seconds per re-derive command
80
- store: /labs/monte-carlo-lab # optional: where finished runs' raw output is published (below)
78
+ store: /labs/monte-carlo-lab # optional: where every run's folder lives once it ends (below)
81
79
  ```
82
80
 
83
81
  Paths in `lab.yaml`, `run.json`, `score.yaml`, and a result's `evidence` and `rederive` fields, are relative to the lab root. folio resolves a result's paths from the charter's `root`, so a lab whose library is in `docs/` sets `root: ..` in `folio.yaml`. Both tools then read the same path the same way.
84
82
 
85
83
  `score.yaml` names its `protocol` and the lock's hash (`lock`), and has three parts. `review` holds the independent reviewer's `verdict` (`pass` or `fail`) and `record`, the path of the file where that verdict is recorded, usually the mission file. `scores` holds one entry per prediction and decision rule the lock names, keyed by its id (`P1`, `D1`): a verdict, the value and the evidence it came from. A prediction's verdict is `HIT`, `MISS` or `INDETERMINATE`; a rule's is `FIRED`, `NOT FIRED` or `INDETERMINATE`. `record` holds one entry per number to record, keyed by a short name, with the result's field names: `title`, `protocol`, `number`, `baseline`, `bound`, `evidence`, `rederive`. The review hands one entry to `record-a-result` as `from: experiments/<slug>/score.yaml#<name>`.
86
84
 
87
- `run.json` holds `protocol`, `lock` (the hash the run was checked against), `config` (the pinned configuration, copied from the protocol), `mission`, `spend`, `command`, `started`, `ended`, `exit`, `pid`, `spent`, `commit` (the git head the run started at; null outside git or before the first commit) and `lab-kit` (the version that started it). Times are UTC, to the second. A run's command finds its folders in the environment: `LAB_RUN_DIR`, `LAB_OUT` and `LAB_WORK`. A run that spends writes `out/spend.json` with `{"spent": <number>}`, in the unit of its mission's cap, and `lab-kit run` copies it into `spent` when the command exits. When the command exits, `lab-kit run` also writes `work.sha256` beside `MANIFEST.sha256`, in the same format, one line per file under `work/`, when that folder exists: the raw output stays out of git, and its hashes go in, so any copy of it can be verified.
85
+ **A run.** Everything a run produces lives in its own folder, `runs/<experiment>/<run id>/`, and never in git. lab-kit fixes two names in it and nothing else: `run.json` and `run.log`. What else a run writes, in which files and folders, is the experiment's choice. A run's command finds its folder in `LAB_RUN` (and, for scripts written before 0.6.0, `LAB_RUN_DIR`, and `LAB_OUT` and `LAB_WORK`, which name `out/` and `work/` inside it).
88
86
 
89
- **The store.** `store:` in `lab.yaml` is optional: an absolute path, such as `/labs/<lab>`, where every machine that runs the lab mounts one shared filesystem. When it is set, a run's raw output leaves the machine that ran it once the run ends: `lab-kit run` publishes `work/` and `run.log` to `<store>/<the run folder's path from the lab root>/`. Publishing copies into `<store>/.incoming/`, checks the copy against `work.sha256` (no file missing, none extra, none changed), renames it into place in one step, makes it read-only, writes `store.json` (`store`, `files`, `published`, `by`) beside `run.json`, and then deletes the local `work/` and `run.log`. A run already in the store is never overwritten: a copy there that verifies counts as published, one that does not is refused. A publish that fails leaves the run where it is, notes why in `run.log`, and never changes the run's exit code; `lab-kit publish --pending` tries again. The command a run starts is never told the store's path, and a lab's instrument never mounts it into the sandbox where the system under test runs: the store holds other runs' raw output, answers included. A machine that does not mount the store sets `LAB_STORE_HOST` to a host that does; lab-kit then copies with rsync over ssh and verifies there with `sha256sum`.
87
+ `run.json` holds `experiment`, `id`, `protocol`, `lock` (the hash the run was checked against), `config` (the pinned configuration, copied from the protocol), `mission`, `spend`, `command`, `host`, `started`, `ended`, `exit`, `pid`, `spent`, `commit` (the git head the run started at; null outside git or before the first commit), `lab-kit` (the version that started it), and, once the run ends, `files` (every file in the folder but run.json, by its path from the folder, with its sha256) and `seal` (one sha256 over the whole record but itself, so neither a file nor the record changes unseen). Times are UTC, to the second. A run that spends writes `spend.json` with `{"spent": <number>}` in its folder, in the unit of its mission's cap, and `lab-kit run` copies it into `spent` when the command exits. A run moved by `lab-kit migrate` also holds `migrated`: its old path, and when.
88
+
89
+ **The store.** `store:` in `lab.yaml` is optional: an absolute path, such as `/labs/<lab>`, where every machine that runs the lab mounts one shared filesystem. With a store, `runs/` is a link to `<store>/runs`, committed as a link. A run is written on local disk, in `.lab/running/<experiment>/<run id>/` (gitignored), so a stalled store never stalls a running command. When the command exits, lab-kit seals the folder and moves it into the store: a copy into `<store>/.incoming/`, checked file by file against the seal (no file missing, none extra, none changed), renamed into `<store>/runs/<experiment>/<run id>/` in one step and made read-only, and only then the local copy deleted. A run already in the store is never overwritten: a copy there that verifies counts as moved, one that does not is refused. A move that fails leaves the run where it is, notes why in `run.log`, and never changes the run's exit code; `lab-kit publish --pending` tries again. Without a store, `runs/` is a folder in the lab and a run is written there directly. The command a run starts is never told the store's path, and a lab's instrument never mounts it into the sandbox where the system under test runs: the store holds other runs' outputs, answers included. A machine that does not mount the store sets `LAB_STORE_HOST` to a host that does; lab-kit then moves runs with rsync over ssh and verifies there with `sha256sum`, and `lab-kit fetch` copies a run back into `.lab/fetched/`, verified. Where `runs/` does not resolve, as in CI, the gate skips the checks that read runs and says so once.
90
90
 
91
91
  `lock.json` holds `protocol`, `path`, `sha256` and `locked`, and from lab-kit 0.3.0 `instrument`, `commit` and `lab-kit`. `instrument` is one sha256 over every file under `experiments/<slug>/bin/` (a `__pycache__` folder left out), sorted by its path from `bin/`, each fed as its path, a newline, then its bytes. `commit` is the git head at the lock (null outside git or before the first commit), and `lab-kit` the version that took it. A lock or a run record without these fields predates them and is read as it is: records are never edited.
92
92
 
93
- A mission file starts with YAML front matter: `title`, `status` (`draft`, `active` or `concluded`), `rests_on` (the ids it rests on), `approved` (who approved the mission, and when, with the date as `YYYY-MM-DD`) and `cap` (a number, in the unit its runs report spend; `0` when nothing spends), and optionally `budget` (the name of a standing budget it draws on). A mission is `draft` until it is approved; `approved` is required from `active` on. Its log is everything under its `## Log` heading; entries are only added at the bottom. `ops/STATE.md` has one line `Active mission: <path>` or `Active mission: none`, and never names a draft.
93
+ A mission file, `missions/<YY-MM-DD>-m_<slug>.md`, starts with YAML front matter: `title`, `status` (`draft`, `active` or `concluded`), `rests_on` (the ids it rests on), `approved` (who approved the mission, and when, with the date as `YYYY-MM-DD`) and `cap` (a number, in the unit its runs report spend; `0` when nothing spends), and optionally `budget` (the name of a standing budget it draws on). A mission is `draft` until it is approved; `approved` is required from `active` on. Its log is everything under its `## Log` heading; entries are only added at the bottom. `ops/STATE.md` has one line `Active mission: missions/<file>.md` or `Active mission: none`, and never names a draft.
94
94
 
95
95
  A standing budget, `ops/budgets/<name>.yaml`, holds `title`, `unit` and `grants`: a list, each grant with `amount` (a positive number, in the unit of its missions' caps) and `approved` (who granted it, and when, with the date as `YYYY-MM-DD`). The budget's amount is the sum of its grants. A mission under it commits its `cap` while it is `active` (or what its runs spent, when that is more) and what its runs spent once it is `concluded`; a draft commits nothing. Its `cap` is a finite number, never negative, and every grant amount and recorded spend is finite. What its missions commit never exceeds its amount. Grants are only added at the bottom. The operator approves a mission alone, or grants a standing budget under which the commander approves missions that fit inside it, as the method says (DISCIPLINE §13, §14).
96
96
 
@@ -105,12 +105,12 @@ question --> protocol (draft) --> review of the draft --> lock --> run --> evide
105
105
  ```
106
106
 
107
107
  1. **Question.** The write skill adds or revises a question `Q-n`. It is ranked, and it says what would settle it.
108
- 2. **Protocol.** The write skill drafts `protocols/{slug}.md`. `lab-kit experiment {slug}` creates `experiments/{slug}/`. The protocol names its question, the intuition, the one variable and its arms, the pinned configuration, the allowed moves, the measures, the decision rules `D1`... with the kill rule marked, and the predictions `P1`... with confidences.
108
+ 2. **Protocol.** The write skill drafts `protocols/{slug}.md`. `lab-kit experiment {slug}` creates `experiments/{YY-MM-DD}-e_{slug}/`. The protocol names its question, the intuition, the one variable and its arms, the pinned configuration, the allowed moves, the measures, the decision rules `D1`... with the kill rule marked, and the predictions `P1`... with confidences.
109
109
  3. **Draft review.** A reviewer reads the draft before the lock. The commander records the verdict in the mission log.
110
110
  4. **Lock.** The write skill sets the protocol's status to `locked`. The protocol's card lists `locked` in `frozen_in`, so folio's `frozen` check holds it from here: nothing changes but its status field. `lab-kit lock {slug}` writes `lock.json` with the file's hash, taken with the status field set aside, so the one change folio allows (to `abandoned`) does not break it. `folio journal add --title .. --description .. --body .. --kind lock --about {slug}` records it. The commit that holds the protocol, `lock.json`, that journal entry and `.folio/` is the lock; folio's `frozen` check compares with the first commit that set `locked`. From here lab-kit also checks the protocol's bytes, and the bytes of the experiment's `bin/`, so the instrument is committed before the lock.
111
- 5. **Run.** `lab-kit run {slug} -- <command>` refuses unless the lock is intact, the instrument included. It writes `run.json`, starts the lab's own command in the background, and logs it. A run under a mission names it with `--mission`, and the mission must be approved and `active`. A run that spends tokens or money also needs a mission that records a cap. Launch, checkpoints and terminations each go in a journal entry of kind `run` about the protocol.
112
- 6. **Evidence.** When the command exits, `lab-kit run` writes `MANIFEST.sha256` and the exit code. `out/` is committed once, at the end, in one commit. A run whose tool failed is invalid, and is recorded as invalid.
113
- 7. **Score.** The review skill reads the lock first. It folds `out/` with a selftested script. `lab-kit score {slug}` creates `score.yaml` with every prediction and rule id the lock names. The review fills each with a verdict, a value and the evidence it came from, and lists the numbers to record under `record`.
111
+ 5. **Run.** `lab-kit run {slug} -- <command>` refuses unless the lock is intact, the instrument included. It creates the run's folder, writes `run.json`, starts the lab's own command in the background, and logs it. A run under a mission names it with `--mission`, and the mission must be approved and `active`. A run that spends tokens or money also needs a mission that records a cap. Launch, checkpoints and terminations each go in a journal entry of kind `run` about the protocol.
112
+ 6. **Evidence.** When the command exits, `lab-kit run` records the exit code and seals the run's folder, then moves it to the store when there is one. Nothing of it goes in git. A run whose tool failed is invalid, and is recorded as invalid.
113
+ 7. **Score.** The review skill reads the lock first. It folds the runs' folders with a tested fold. `lab-kit score {slug}` creates `score.yaml` with every prediction and rule id the lock names. The review fills each with a verdict, a value and the evidence it came from, and lists the numbers to record under `record`.
114
114
  8. **Reviewer pass.** An independent reviewer re-derives the scored numbers and returns a verdict. The commander records it, and `score.yaml` points to it.
115
115
  9. **Result.** For each number worth citing, folio's run skill follows `record-a-result` with `from: experiments/{slug}/score.yaml#<name>`, and writes `R-n`. Its number, baseline, bound, evidence path, re-derive command and protocol come from that entry; its date is the day it is recorded. A result is permanent and writes no status. folio derives it: `live` until a newer result supersedes it or a retraction names it. A null or a fired kill rule is a result too.
116
116
  10. **Claim.** If the lab wants to say something, the write skill adds or revises a claim `C-n` citing its results.
@@ -130,26 +130,25 @@ A result that is not promoted is recorded too: a journal entry about the protoco
130
130
  | --- | --- |
131
131
  | `lab-lock-recorded` | Every protocol with status `locked` has a `lock.json` naming it, and every `lock.json` names a protocol whose status is `locked` or `abandoned`. |
132
132
  | `lab-lock-intact` | A locked protocol's bytes, with its status field set aside, hash to the value in its `lock.json`. Nothing but the status changes after the lock. When the lock records `instrument`, the experiment's `bin/` hashes to it. A lock without `instrument` predates it and passes. |
133
- | `lab-run-after-lock` | Every run belongs to a locked experiment, records the lock's hash, and started after the lock date. |
133
+ | `lab-run-after-lock` | Every run belongs to an experiment that exists and is locked, records the lock's hash, and started after the lock date. |
134
134
  | `lab-roster-frozen` | Every run's configuration in `run.json` equals the `yaml` block under its locked protocol's `## Pinned configuration`. |
135
- | `lab-evidence-sealed` | Every finished run's `out/` matches its `MANIFEST.sha256`. A run with no end and no live process is reported as orphaned (warning). |
135
+ | `lab-evidence-sealed` | Every ended run's folder matches the seal in its `run.json`: no file missing, none extra, none changed. A run with no end and no live process is reported as orphaned (warning). Where `runs/` does not resolve, it says once that the checks reading runs were skipped (warning). |
136
136
  | `lab-rederive` | Every result whose derived status is `live` has a `rederive` command that, run from the lab root within the timeout, exits 0 and prints the result's `number` exactly, after trimming whitespace. |
137
- | `lab-result-grounded` | Every result whose derived status is `live` names a locked protocol. Its evidence path is inside a finished run of that experiment. The experiment's score records a reviewer pass. |
137
+ | `lab-result-grounded` | Every result whose derived status is `live` names a locked protocol. Its evidence path is inside a finished run of that experiment, `runs/<experiment>/<run id>/...`, and exists. The experiment's score records a reviewer pass. |
138
138
  | `lab-score-exact` | A `score.yaml` names the lock's hash. Its `scores` list exactly the locked protocol's prediction ids (`P1`...) and decision-rule ids (`D1`...), no more and no fewer. Each has an allowed verdict and an existing evidence path. |
139
139
  | `lab-ids-resolve` | Every id cited in a lab file (state, missions, lock records, runs, scores) resolves to a document in the library. |
140
140
  | `lab-frozen-intact` | Every surface listed under `frozen:` matches its hash in `.lab/frozen.sha256`. |
141
- | `lab-selftest` | Every script in `experiments/*/bin/` and under `tools:` passes `--selftest`. |
142
141
  | `lab-mission-approved` | Every mission file has status `draft`, `active` or `concluded`. An `active` or `concluded` mission records `approved`, naming who and a date. No run names a draft mission, and `ops/STATE.md` names no draft as the active mission. Every standing budget has grants, each with a positive `amount` and an `approved` naming who and a date. A mission's `budget` names a budget that exists; past draft, such a mission records a numeric `cap`; and what the missions under a budget commit fits inside its amount. |
143
142
  | `lab-spend-recorded` | A run marked as spending names a mission that records the operator's approval and a cap, and its recorded spend is within the cap. |
144
- | `lab-append-only` | Lab files only: committed mission log entries, standing budget grants and units, lock records and run records are unchanged in later commits; a mission file or a budget is never deleted. A run record committed while its run was going may later fill in `ended`, `exit` and `spent`, each once: nothing else in it changes, and a closing field once set never changes again. Journal entries and results are folio's `permanent` check. |
143
+ | `lab-append-only` | Lab files only: committed mission log entries, standing budget grants and units, and lock records are unchanged in later commits; a mission file or a budget is never deleted. A file `lab-kit migrate` moved, listed in `.lab/migration.tsv`, is read at its new path. Runs are never in git; their seal holds them. Journal entries and results are folio's `permanent` check. |
145
144
  | `lab-method-current` | `.lab/method/`, the five skills and the four roles match the lab-kit version in `lab.yaml`. |
146
145
  | `lab-state-pointer` | `ops/STATE.md` names an active mission file or "none", stays under 300 words, and cites ids for any number (warning). |
147
146
  | `lab-scored-reported` | Every scored experiment has a report citing its results, or a journal entry about its protocol saying why not (warning). |
148
- | `lab-work-stored` | When `lab.yaml` names a `store:`, every ended run with a `work.sha256` has a `store.json`: its raw output was published (warning). |
147
+ | `lab-run-stored` | When `lab.yaml` names a `store:`, no ended run is left on this machine in `.lab/running/` (warning). |
149
148
 
150
- The gate checks that the record is consistent. It does not check that the record is honest. That is the reviewer's job.
149
+ The gate checks that the record is consistent. It does not check that the record is honest. That is the reviewer's job. Nor does it test the lab's code: the lab's own gate runs the lab's test suite beside `lab-kit check`, and an experiment's command calls that tested code.
151
150
 
152
- lab-kit runs every selftest, re-derive command and run from the lab root with `PYTHONDONTWRITEBYTECODE=1`, so no bytecode lands in a frozen surface. A run counts as ended for `lab-evidence-sealed` whatever its exit code; for `lab-result-grounded` it must have exited 0.
151
+ lab-kit runs every re-derive command and run from the lab root with `PYTHONDONTWRITEBYTECODE=1`, so no bytecode lands in a frozen surface. A run counts as ended for `lab-evidence-sealed` whatever its exit code; for `lab-result-grounded` it must have exited 0.
153
152
 
154
153
  ## 6. The `lab-kit` command
155
154
 
@@ -157,15 +156,16 @@ Only skills call it, the way an agent calls git. Every command that changes file
157
156
 
158
157
  | Command | Does |
159
158
  | --- | --- |
160
- | `lab-kit init [--library <dir>]` | Writes `lab.yaml`, `ops/`, `experiments/`, `.lab/`, the five skills and the four roles, and links `.claude/skills` to `.agents/skills` and `.claude/agents` to `.agents/agents`. With no `--library`, no `lab.yaml` and no `folio.yaml` at the root, it installs only the method, the skills and the roles, so the set-up-lab skill can be read before the library exists. With `--library`, the folder must hold a `folio.yaml`. In the library it only switches the lab pack on, if set-up has not. It refuses a library whose `root` is not the lab root. Run again in a lab, it restores the method, the skills and the roles to this version's, and sets `lab-kit:` in `lab.yaml` to it. |
159
+ | `lab-kit init [--library <dir>]` | Writes `lab.yaml`, `missions/`, `ops/`, `experiments/`, `runs/` (a link to the store when there is one), `.lab/`, the five skills and the four roles, and links `.claude/skills` to `.agents/skills` and `.claude/agents` to `.agents/agents`. With no `--library`, no `lab.yaml` and no `folio.yaml` at the root, it installs only the method, the skills and the roles, so the set-up-lab skill can be read before the library exists. With `--library`, the folder must hold a `folio.yaml`. In the library it only switches the lab pack on, if set-up has not. It refuses a library whose `root` is not the lab root. Run again in a lab, it restores the method, the skills and the roles to this version's, and sets `lab-kit:` in `lab.yaml` to it. |
161
160
  | `lab-kit check [--only <id>] [--json]` | The lab gate: `folio check`, then the checks above. `--only` runs one check, and may repeat; `folio` names folio's gate. Exit 0 only when there is no error. |
162
- | `lab-kit experiment <slug>` | Creates `experiments/<slug>/` with `bin/`, `runs/` and the gitignore for scratch. |
161
+ | `lab-kit experiment <slug>` | Creates `experiments/<YY-MM-DD>-e_<slug>/` with `bin/`. |
163
162
  | `lab-kit lock <slug>` | Writes `lock.json` for a protocol whose status is `locked` and whose gate is green, hashing the file with its status field set aside, and records the instrument's hash, the git commit and the lab-kit version. Refuses otherwise. |
164
- | `lab-kit run <slug> [--mission <file>] [--spend] [--wait] -- <command>` | Checks the lock, writes `run.json`, starts the command in the background from the lab root, records the git commit and the lab-kit version, and writes `MANIFEST.sha256` for `out/` and `work.sha256` for `work/` when it exits. It refuses when the protocol or `bin/` no longer hashes to the lock. With `--mission`, it refuses a mission whose status is not `active` or that records no approval. `--wait` waits for the end, for a short token-free stage. |
163
+ | `lab-kit run <slug> [--mission <file>] [--spend] [--wait] -- <command>` | Checks the lock, creates the run's folder, writes `run.json`, starts the command in the background from the lab root, records the git commit, the host and the lab-kit version, and when it exits seals the folder and moves it to the store. It refuses when the protocol or `bin/` no longer hashes to the lock. With `--mission`, it refuses a mission whose status is not `active` or that records no approval. `--wait` waits for the end, for a short token-free stage. |
165
164
  | `lab-kit runs [--live]` | Lists runs and their state: running, finished, failed, orphaned. |
166
- | `lab-kit publish <slug>/<run id>... \| --pending [--keep]` | Publishes ended runs' raw output to the store, as above. `--pending` takes every ended run whose `work/` is still here. `--keep` keeps the local copy. |
167
- | `lab-kit fetch <slug>/<run id>` | Copies a published run's `work/` and `run.log` back into its folder, writable, and verifies them against `work.sha256`. |
168
- | `lab-kit verify <slug>/<run id>` | Checks the store's copy of a run against its `work.sha256`. Changes nothing. |
165
+ | `lab-kit publish <experiment>/<run id>... \| --pending [--keep]` | Moves ended runs from `.lab/running/` into the store, as above. `--pending` takes every ended run still here. `--keep` keeps the local copy. |
166
+ | `lab-kit fetch <experiment>/<run id>` | On a machine without the store, copies a run from `LAB_STORE_HOST` into `.lab/fetched/` and verifies it against its seal. |
167
+ | `lab-kit verify <experiment>/<run id>` | Checks a run's folder against its seal, here or on `LAB_STORE_HOST`. Changes nothing. |
168
+ | `lab-kit migrate [--apply] [--orphaned <path>]` | Moves a lab from the layout before 0.6.0 to this one, in two phases so a failure loses nothing. First every run is assembled (in the store's `.incoming/` when there is one), verified against its old manifests, sealed and put in place: one folder per run in `runs/`, its `out/`, `work/` and `run.log` together. Then git: missions to `missions/`, experiments to their dated ids, every old path and id in git-tracked text rewritten (never inside a longer name or a file name), the lock hashes taken again where the rewrite changed a protocol or `bin/`, the rewritten documents listed in the library's `baseline.sha256` (folio's `baseline`), and every old name kept in `.lab/migration.tsv`. It refuses to start unless folio's `permanent` and `frozen` and `lab-lock-intact` pass, so the rewrite hides no earlier edit, and it moves a run with no end only when `--orphaned` names it. A retry after a failure reuses the runs already in place. A dry run without `--apply`. |
169
169
  | `lab-kit score <slug>` | Creates `score.yaml` from the locked protocol's prediction and rule ids, with empty verdicts. |
170
170
  | `lab-kit rederive <R-n>` | Runs one result's re-derive command and compares its output with the number. |
171
171
  | `lab-kit freeze <path>` | Records a frozen surface's hash. Only the operator asks for this. |
@@ -0,0 +1,41 @@
1
+ ---
2
+ name: runner
3
+ description: >-
4
+ Executes a fully specified pipeline for a locked protocol: builds workspaces, invokes tools and
5
+ containers, launches runs with lab-kit, recovers pinned artefacts, folds evidence, and scores
6
+ predictions neutrally. Use when the design is done (a locked protocol or a complete brief
7
+ exists) and what remains is disciplined execution. Not for designing protocols or interpreting
8
+ results.
9
+ tools: Read, Grep, Glob, Bash, Write, Edit
10
+ ---
11
+
12
+ You execute measurement pipelines. The design is done. Your job is disciplined, fail-loud execution and an accurate account.
13
+
14
+ Execution:
15
+
16
+ - Work only from a locked protocol or a complete brief. If the protocol is not locked, stop and say so.
17
+ - Launch every run with `lab-kit run`. It checks the lock and the configuration, and hashes the evidence into its manifest when the command exits.
18
+ - Copy the mechanics of the earlier experiment the protocol names: its scripts, its pins, its gitignore. Do not invent new ones.
19
+ - Verify every recovered artefact against its expected hash. Report a mismatch; never substitute silently.
20
+ - Write only in the experiment's `bin/` and in the run's own folder (`$LAB_RUN`). Evidence reaches it through the run's own command.
21
+ - Fail loud. An apparatus failure is a result to report, not a thing to patch around.
22
+
23
+ Reporting:
24
+
25
+ - Report once, at completion, with the full account the brief asks for. No step-by-step narration.
26
+ - To wait on long work, arm one watcher that fires only on the end marker or an error signature. Silence between launch and end is correct.
27
+ - List every deviation and surprise with its time, so the experiment skill can journal it.
28
+
29
+ Hard limits:
30
+
31
+ - No commits. No documents: you never write in the library, the journal included.
32
+ - Never edit a locked protocol, a finished run, a lock record or a frozen surface.
33
+ - Never change the configuration the protocol pins, and never raise a budget cap.
34
+ - Never read a held-out or sealed set into a workspace you build.
35
+ - Respect the concurrency cap in the brief. On a shared machine, default to modest parallelism.
36
+
37
+ Honesty:
38
+
39
+ - Score predictions neutrally: hits, misses and indeterminates alike.
40
+ - A miss or a null is a full result.
41
+ - Never word a claim more strongly than the brief allows.
@@ -30,7 +30,7 @@ You take one question from a draft protocol to a finished run with its evidence
30
30
  7. Fail loud. An arm whose tool fails is invalid, never downgraded. Undecidable is not a pass. Judged, not checked.
31
31
  8. Name every move the agent under test may make, under the protocol's allowed moves. A move the list does not name is one it may never use. The section is checked by the protocol genre's required parts; its completeness is judged.
32
32
  9. No live spend without the operator's word, recorded in a mission with a cap. Checked by `lab-spend-recorded`.
33
- 10. Every script in `bin/` has a `--selftest`. Checked by `lab-selftest`.
33
+ 10. The code a run executes is tested by the lab's own test suite. Prefer a command that calls the lab's own tested code (its package or its tools) over a script written for one experiment; keep in `bin/` only what is particular to this experiment, and test it there too. Judged, not checked.
34
34
  11. Evidence is committed once, at the end of the run. Scratch and logs stay out of git. Checked by `lab-evidence-sealed`; the rest is judged.
35
35
  12. Every document goes through the write skill. You never write one by hand. Checked by `folio check`.
36
36
 
@@ -45,14 +45,14 @@ You take one question from a draft protocol to a finished run with its evidence
45
45
  - the decision rules `D1`..., with the kill rule marked `(kill)`;
46
46
  - the predictions `P1`..., each with a confidence;
47
47
  - what it cannot show.
48
- 2. **Make its folder.** Run `lab-kit experiment <slug>`. Put the lab's scripts in `experiments/<slug>/bin/`, each with a `--selftest`. Run `lab-kit check --only lab-selftest`.
48
+ 2. **Make its folder.** Run `lab-kit experiment <slug>`. The run's command calls the lab's own tested code; anything particular to this experiment goes in `experiments/<YY-MM-DD>-e_<slug>/bin/`, with its tests in the lab's suite. An experiment that needs nothing of its own leaves `bin/` empty. Run the lab's tests.
49
49
  3. **Review the draft.** Send a reviewer with fresh context. Fold its points into the draft through the write skill. A protocol is cheap to fix before the lock and impossible to fix after.
50
- 4. **Lock.** Commit `bin/` first: the lock hashes it, and `lab-lock-intact` fails if it changes after. Through the write skill, set the protocol's status to `locked`. Run `lab-kit lock <slug>`. Record it: `folio journal add --title .. --description .. --body .. --kind lock --about <slug>,<Q-n>`. Run `folio index`, then `lab-kit check`. Commit exactly the protocol, `lock.json`, the `lock` journal entry and `.folio/` together; that commit is the lock. From here folio's `frozen` check and `lab-lock-intact` hold it.
50
+ 4. **Lock.** Commit `bin/` and the code the command calls first: the lock hashes `bin/` and records the commit, and `lab-lock-intact` fails if `bin/` changes after. The lab's code outside `bin/` may change later; the lock's commit says which version the run used, so a protocol that needs it fixed pins it by commit. Through the write skill, set the protocol's status to `locked`. Run `lab-kit lock <slug>`. Record it: `folio journal add --title .. --description .. --body .. --kind lock --about <slug>,<Q-n>`. Run `folio index`, then `lab-kit check`. Commit exactly the protocol, `lock.json`, the `lock` journal entry and `.folio/` together; that commit is the lock. From here folio's `frozen` check and `lab-lock-intact` hold it.
51
51
  5. **De-risk.** Run the token-free stages first, with `lab-kit run`. Fix the apparatus only before the lock. After it, an apparatus fault ends the run as invalid.
52
52
  6. **Launch.** Run `lab-kit run <slug> -- <command>`. Add `--spend --mission <file>` for a run that spends. Watch it at once: arm one watcher on its log that fires on the end marker or an error signature. A silent launch failure can waste a whole night.
53
53
  7. **Supervise.** Write a journal entry of kind `run` about the protocol for each launch, checkpoint, surprise and termination, with `folio journal add`. Keep each entry to what changed. Record a deviation the moment you see it. Commit each entry with `.folio/` after `folio index`; never stage the run's working files.
54
54
  8. **On failure.** End the run. Start a fresh one with the same configuration. Disclose it in a `run` entry. Never swap a model or a tool mid-run.
55
- 9. **Conclude.** When the run ends, `lab-kit run` writes the manifest of `out/`. Commit `out/`, `MANIFEST.sha256`, `work.sha256` and `run.json` in one commit, with `store.json` when `lab.yaml` names a store; `work/` stays out of git, and `work.sha256` is what any copy of it is verified against. With a store, `lab-kit run` has already published `work/` there and deleted the local copy; if `run.log` says the publish failed, run `lab-kit publish --pending`. Never mount the store where the system under test runs: it holds other runs' answers. Run `lab-kit check`. Hand the experiment to the review skill. Do not score your own run.
55
+ 9. **Conclude.** When the run ends, `lab-kit run` seals its folder (every file's hash goes in its `run.json`) and, when `lab.yaml` names a store, moves it to `runs/<experiment>/<run id>/` there, verified, and deletes the local copy; if `run.log` says the move failed, run `lab-kit publish --pending`. Nothing of a run goes in git: results and scorecards cite it by its path under `runs/`. Never mount the store where the system under test runs: it holds other runs' answers. Run `lab-kit check`. Hand the experiment to the review skill. Do not score your own run.
56
56
 
57
57
  ## Stops
58
58
 
@@ -38,7 +38,7 @@ You plan the mission with the operator, and you stop at their approval. The run-
38
38
  - the spend: token-free, or which runs may spend and the cap, as a number in one unit, and the standing budget it draws on, if any (`lab-kit status` shows each budget and the missions under it);
39
39
  - what must never be touched: frozen surfaces, locked protocols, recorded results, beyond what `lab.yaml` already freezes;
40
40
  - the stops: where the run must come back to the operator, beyond the method's own.
41
- 3. **Draft.** Write `ops/missions/{date}-{slug}.md` with front matter `title`, `status: draft`, `rests_on` (the ids it serves), `cap` (a number; `0` for a token-free mission), and `budget` (the standing budget's name) when it draws on one. Its body holds these sections, in order:
41
+ 3. **Draft.** Write `missions/{YY-MM-DD}-m_{slug}.md` with front matter `title`, `status: draft`, `rests_on` (the ids it serves), `cap` (a number; `0` for a token-free mission), and `budget` (the standing budget's name) when it draws on one. Its body holds these sections, in order:
42
42
  - `## Objective`: the operator's words;
43
43
  - `## Rests on`: the ids, and why;
44
44
  - `## Done when`: the milestones, each with its observable;
@@ -27,7 +27,7 @@ You turn a finished experiment's evidence into a scorecard against its lock, the
27
27
  3. Score exactly what the lock names: every prediction and every rule, no more and no fewer. Checked by `lab-score-exact`.
28
28
  4. A badly chosen rule is still the rule. Record the flaw as a lesson; never rescore. Judged, not checked.
29
29
  5. A miss is a result. Nulls, fired kill rules and invalid arms are recorded at the same length as wins. Checked in part by `lab-scored-reported`; the rest is judged.
30
- 6. Every fold script has a `--selftest` on a planted case with a known answer. Checked by `lab-selftest`.
30
+ 6. Every fold is tested against a planted case with a known answer, in the lab's own test suite. Judged, not checked.
31
31
  7. You score; the reviewer gates. Nothing is recorded as a result before a reviewer pass. Checked by `lab-result-grounded`.
32
32
  8. Every document goes through folio: results and reports through workflows that folio's run skill follows, the rest through the write skill. Checked by `folio check`.
33
33
  9. A report is frozen once `live`, and a result is permanent. A new result earns a new report, never a rewrite of an old one. folio shows a banner on the old report when a result it cites is superseded or retracted. Checked by folio's `frozen` and `permanent` checks.
@@ -36,7 +36,7 @@ You turn a finished experiment's evidence into a scorecard against its lock, the
36
36
 
37
37
  ## Steps
38
38
 
39
- 1. **Fold.** Run the experiment's fold script over each finished run's `out/`. If none exists, write one in `bin/` with a `--selftest` on a planted case, and land it as an instrument change. Never fold by eye.
39
+ 1. **Fold.** Run the experiment's fold script over each finished run's folder under `runs/<experiment>/`. If none exists, write one, with a test on a planted case in the lab's suite, and land it as an instrument change. Never fold by eye.
40
40
  2. **Score.** Run `lab-kit score <slug>`. Under `scores`, for every prediction `P1`..., fill HIT, MISS or INDETERMINATE, with the value and the evidence file it came from. For every decision rule `D1`..., say whether it fired, the kill rule included. Run `lab-kit check --only lab-score-exact`.
41
41
  3. **Gate.** Ask the commander for a reviewer pass, or send one yourself if you are the commander. The reviewer re-derives the scored numbers. Record its verdict where the mission keeps verdicts, and point `score.yaml` to it: `review.verdict` and `review.record`.
42
42
  4. **Choose what to record.** For each scored number, ask: is this a citable fact? A null counts. Add each one to `score.yaml` under `record`, keyed by a short name. Give it the result's fields: `title`, `protocol`, `number`, `baseline`, `bound`, `evidence` and `rederive`. The `number` is exactly what the re-derive command prints, and the evidence lies inside the run. Name every number you leave out.
@@ -41,7 +41,7 @@ Create the lab's library and files, record its question, and leave `lab-kit chec
41
41
  - `root`: `..` when the library is in `docs/`, nothing when it is the repository root.
42
42
 
43
43
  Skip its journal entry and its commit: step 10 records and commits the whole setup once.
44
- 4. **Install the lab.** Run `lab-kit init --library <dir>`. It writes `lab.yaml`, `ops/STATE.md`, `ops/missions/`, `experiments/` and `.lab/`, restores the method, the skills and the roles to this version, links `.claude/agents` to `.agents/agents`, and switches the lab pack on if set-up did not.
44
+ 4. **Install the lab.** Run `lab-kit init --library <dir>`. It writes `lab.yaml`, `ops/STATE.md`, `missions/`, `experiments/`, `runs/` (a link to the store when `lab.yaml` names one) and `.lab/`, restores the method, the skills and the roles to this version, links `.claude/agents` to `.agents/agents`, and switches the lab pack on if set-up did not.
45
45
  5. **Add the front door.** Through folio's write skill, add a project document `lab`. Its description states the question. Its `reviewed` date is today, its state says the lab is set up and no mission is active, and its next step is the operator's first objective. Create a top-level map, `research`, through the write skill, keep it `live`, add the front door to it, and set it on the home page: `folio config set home.maps research`. Set `front: lab` in `lab.yaml`.
46
46
  6. **Add the question.** Through the write skill, add the question as `Q-1`, with rank 1, why it matters, what would settle it, and what would make the lab drop it. Add it to the `research` map.
47
47
  7. **Freeze the surfaces.** List each path the operator named under `frozen:` in `lab.yaml`, then run `lab-kit freeze <path>` for each. With none, leave `frozen: []`.
@@ -0,0 +1,4 @@
1
+ __pycache__/
2
+ # lab-kit: a run's folder while it runs here, and runs fetched from the store
3
+ /.lab/running/
4
+ /.lab/fetched/
@@ -55,7 +55,7 @@ No defensive fallbacks. No silent degradation. An arm whose tool fails is invali
55
55
 
56
56
  An apparatus failure is a result to report, not a thing to patch around. Judged, not checked.
57
57
 
58
- Every script ships a `--selftest`, and the gate runs it. Checked by `lab-selftest`.
58
+ Every piece of code a run executes is tested by the lab's own test suite, and the lab's gate runs it. lab-kit runs no lab code in its checks: it checks the record, the lab tests its code. Judged, not checked.
59
59
 
60
60
  Every fold is tested against a planted case with a known answer. A fold that looks fine can still be blind to a renamed field. Judged, not checked.
61
61
 
@@ -12,7 +12,7 @@ evidence -> protocol -> journal -> result -> claim -+-> report (frozen once liv
12
12
 
13
13
  | Level | Home | Owns | Changes |
14
14
  | --- | --- | --- | --- |
15
- | Evidence | `experiments/{slug}/runs/{run-id}/out/` (lab file) | the raw measurement | never; hashed in its manifest (`lab-evidence-sealed`) |
15
+ | Evidence | `runs/{experiment}/{run-id}/` (outside git, in the lab's store) | the raw measurement | never; sealed by the hash of every file in its `run.json` (`lab-evidence-sealed`) |
16
16
  | Protocol | `content/protocols/{slug}.md` (document) | the plan: one variable, measures, rules, kill rule, predictions | frozen at lock: status `locked` (folio's `frozen`, `lab-lock-intact`) |
17
17
  | Score | `experiments/{slug}/score.yaml` (lab file) | each prediction and rule, scored against the lock | written once, at the review (`lab-score-exact`) |
18
18
  | Journal | `content/journal/{yyyy}/{date}-{slug}.md`, one file per entry (document) | what happened, what it means, what went wrong | never edited; a new entry each time (folio's `permanent`) |
@@ -23,7 +23,7 @@ evidence -> protocol -> journal -> result -> claim -+-> report (frozen once liv
23
23
  | Page | the library's maps, concepts and front door (documents) | the living explanation | updated when a result changes |
24
24
  | Paper | the library's papers (documents) | the venue article | frozen at submission (folio's `frozen`) |
25
25
  | State | `ops/STATE.md` (lab file) | what is true now and what is next | a pointer, rewritten freely (`lab-state-pointer`) |
26
- | Missions | `ops/missions/` (lab files) | one operator objective, its approved plan and its log | approved before it runs (`lab-mission-approved`); the log is append-only (`lab-append-only`) |
26
+ | Missions | `missions/` (lab files) | one operator objective, its approved plan and its log | approved before it runs (`lab-mission-approved`); the log is append-only (`lab-append-only`) |
27
27
  | Standing budgets | `ops/budgets/` (lab files) | money the operator granted once for many missions | the missions under it fit inside it (`lab-mission-approved`); grants are only added (`lab-append-only`) |
28
28
 
29
29
  ## Promotion
@@ -0,0 +1,4 @@
1
+ # kind old new
2
+ mission ops/missions/2026-10-04-pi-error-scaling.md missions/26-10-04-m_pi-error-scaling.md
3
+ experiment pi-error-scaling 26-10-04-e_pi-error-scaling
4
+ run pi-error-scaling/20261004-231629 26-10-04-e_pi-error-scaling/26-10-04-2316-r_zlk7