Issues

• repo tasks: authoring replay tasks from merged PRs

  intel/dokimi#294 · Filter by author antonwolfy opened on Oct 29, 2026

• novelty: spread the atom cap across the artifact instead of its opening

  intel/dokimi#293 · Filter by author antonwolfy opened on Oct 29, 2026

• repo compare: an interval that agrees with the sign-flip p

  intel/dokimi#292 · Filter by author antonwolfy opened on Oct 29, 2026

• review_replay: optional judged recall over the credited hits

  intel/dokimi#291 · Filter by author antonwolfy opened on Oct 29, 2026

• repo compare: flag replay tasks a guidance revision may have seen

  intel/dokimi#290 · Filter by author antonwolfy opened on Oct 29, 2026

• runs list scans the runs root; index.jsonl is a cache (#94 Q6)

  intel/dokimi#278 · Filter by author napetrov opened on Oct 29, 2026
  

• Artifact envelope for run artifacts, not inputs (#94 B1/B1b)

  intel/dokimi#277 · Filter by author napetrov opened on Oct 29, 2026

• CLI exit codes: audit every command against the 0/1/2 rule (#94 A2)

  intel/dokimi#275 · Filter by author napetrov opened on Oct 29, 2026

• CLI: move commands into the new groups, stage M1–M3 (#94)

  intel/dokimi#276 · Filter by author napetrov opened on Oct 29, 2026

• Consolidate duplicated subsystems (stats, LLM routing, judge loop, containers, digests, writers)

  intel/dokimi#229 · Filter by author napetrov opened on Oct 26, 2026
  1

• Answering stack: one implementation for evaluate/benchmark/answers and arms/subjects

  intel/dokimi#228 · Filter by author napetrov opened on Oct 26, 2026
  1

• Skill auto-improvement: measurement gates (A) and improve command (B)

  intel/dokimi#220 · Filter by author napetrov opened on Oct 25, 2026

• novelty: bad LLM replies cached or misreported (judge retry, conflict silent drop, retry warning) — follow-up to #40

  intel/dokimi#219 · Filter by author napetrov opened on Oct 25, 2026

• repo run: block fork-context skills under --delegation none

  intel/dokimi#148 · Filter by author napetrov opened on Oct 21, 2026

• Skill catalog integration (intel/skills): per-skill evals in catalog CI, results, promotion to validated

  intel/dokimi#138 · Filter by author napetrov opened on Oct 20, 2026
  1

• User loop: eval engine contracts (plan/validate/compare/gate/record), 3 skills, GitHub Action

  intel/dokimi#121 · Filter by author napetrov opened on Oct 19, 2026
  1

• Journeys: bind available but untraced criteria to tests (AC-J03-002, AC-J03-005, AC-J12-001, AC-J14-005)

  intel/dokimi#100 · Filter by author napetrov opened on Oct 19, 2026

• J11/J12: distinct inconclusive verdict and blocking policy (AC-J11-002, AC-J12-002)

  intel/dokimi#99 · Filter by author napetrov opened on Oct 19, 2026

• J2: join the novelty diagnosis to a task-level effect (AC-J02-004, AC-J02-005)

  intel/dokimi#98 · Filter by author napetrov opened on Oct 19, 2026

• J7: draft tasks, negative controls and coverage gaps (AC-J07-004, AC-J07-006, AC-J07-007)

  intel/dokimi#97 · Filter by author napetrov opened on Oct 19, 2026

• J4: adherence and task-impact evidence for instructions (AC-J04-005, AC-J04-006)

  intel/dokimi#96 · Filter by author napetrov opened on Oct 19, 2026

• User journeys J1–J14: catalog and criterion-level evidence tracking

  intel/dokimi#95 · Filter by author napetrov opened on Oct 19, 2026

• CLI API: the evaluation loop at the top level, subject and eval groups below, one enforced flag vocabulary (proposal v2.1)

  enhancement
  intel/dokimi#94 · Filter by author napetrov opened on Oct 19, 2026
  2

• build: hash the dependency lock and install it with --require-hashes

  intel/dokimi#61 · Filter by author napetrov opened on Oct 15, 2026

• judge-validity: validate the judge in the setting arms are scored — persist judge material by digest

  intel/dokimi#48 · Filter by author napetrov opened on Oct 15, 2026