Issues

• feat(desktop): run picker to reopen and compare an application's runs

  akoita/draft-loop#1176 · Filter by author akoita opened on Oct 11, 2026 · Tailored, sendable CVs

• feat(desktop): show what the author received and which job items got no evidence

  akoita/draft-loop#1175 · Filter by author akoita opened on Oct 11, 2026 · Tailored, sendable CVs

• Observe a full-source frontier run on the Euronext reference application

  akoita/draft-loop#1172 · Filter by author akoita opened on Oct 11, 2026 · Tailored, sendable CVs

• feat(evaluations): stop on finding resolution, pause on persisting findings

  akoita/draft-loop#1113 · Filter by author akoita opened on Oct 9, 2026

• feat(application): targeted patch revisions for accepted findings

  akoita/draft-loop#1112 · Filter by author akoita opened on Oct 9, 2026

• feat(application): bounded validate_proposal and search_evidence tools for the API author

  akoita/draft-loop#1111 · Filter by author akoita opened on Oct 9, 2026

• feat(desktop): plain-words reason the loop stopped, next to Approve

  akoita/draft-loop#1110 · Filter by author akoita opened on Oct 9, 2026

• feat(desktop): run timeline and score trajectory for one run

  akoita/draft-loop#1109 · Filter by author akoita opened on Oct 9, 2026

• feat(orchestrator): attribute each revised block to the findings it addresses

  akoita/draft-loop#1108 · Filter by author akoita opened on Oct 9, 2026

• feat(evaluations): local live eval suite on Evalite, run like release:preflight

  akoita/draft-loop#1107 · Filter by author akoita opened on Oct 9, 2026

• feat(evaluations): critic agreement report against labelled failures

  akoita/draft-loop#1106 · Filter by author akoita opened on Oct 9, 2026

• docs(evaluation): label synthetic-cohort rounds into a binary failure taxonomy

  akoita/draft-loop#1105 · Filter by author akoita opened on Oct 9, 2026

• feat(application): opt-in localhost OTLP export of run spans for development

  akoita/draft-loop#1104 · Filter by author akoita opened on Oct 9, 2026

• feat(storage): persist run spans beside run history

  akoita/draft-loop#1103 · Filter by author akoita opened on Oct 9, 2026

• feat(orchestrator): content-free tracing port with GenAI spans per run, step and attempt

  akoita/draft-loop#1102 · Filter by author akoita opened on Oct 9, 2026

• Harness observability, evaluation and loop (rollup)

  akoita/draft-loop#1101 · Filter by author akoita opened on Oct 9, 2026

• Simple guided workflow (rollup)

  akoita/draft-loop#1092 · Filter by author akoita opened on Oct 9, 2026 · Simple guided workflow

• docs(desktop): walk the first-run journey end to end and record it as the stage evidence

  akoita/draft-loop#1091 · Filter by author akoita opened on Oct 9, 2026 · Simple guided workflow

• feat(desktop): plain-language Career profile page — status first, history and checksums under Details

  akoita/draft-loop#1090 · Filter by author akoita opened on Oct 9, 2026 · Simple guided workflow

• feat(desktop): one path to a new application — merge the New application flow into the application steps

  akoita/draft-loop#1089 · Filter by author akoita opened on Oct 9, 2026 · Simple guided workflow

• feat(desktop): the application screen as three steps — job & requirements, profile, review

  akoita/draft-loop#1088 · Filter by author akoita opened on Oct 9, 2026 · Simple guided workflow

• feat(desktop): explain every start blocker next to the Start button, with the action that fixes it

  akoita/draft-loop#1087 · Filter by author akoita opened on Oct 9, 2026 · Simple guided workflow

• Guided career interview, with local voice input (rollup)

  akoita/draft-loop#1030 · Filter by author akoita opened on Oct 8, 2026 · Guided career interview

• ci(desktop): fail packaging when a Vite main/preload build fails

  akoita/draft-loop#1009 · Filter by author akoita opened on Oct 7, 2026

• Unprofiled Gemini requests on large inputs time out before the first stream chunk

  akoita/draft-loop#911 · Filter by author akoita opened on Oct 6, 2026 · Manual parity: inputs