github awslabs/aidlc-workflows v2.11.1-preview.20261012.1
AI-DLC Workflow 2.11.1-preview.20261012.1
on GitHub

pre-release3 hours ago

Warning: Full Suite failed for this source. A Full Suite failure report ends these notes.

Merged commits since preview source d484df9:

  • fix(engine): a setting kept in a chat with no work selected goes to the work it picks up, not only to new work (#2381)
  • fix(install): config leaves a root file that is not UTF-8 untouched and names it, as session start does (#2377)
  • fix(plan-approval): the person says "build it anyway" in their own words; no sentence to type (#2380)
  • fix(engine): a plan typed alone asks what the work is for instead of starting it with a placeholder (#2378)
  • fix(answers): the answer-mode question is known by its own field, so a stage's question that starts the same way is answered as asked (#2379)
  • chore(docs): the learning-loop worked example uses a neutral banking project (#2376)
  • fix(guard): /aidlc --config reads and lands a setting while a plan waits for approval (#2375)
  • fix(engine): a setting the person types is recorded whole and applied once by the setter the agent runs, and stays theirs (#2372)
  • fix(replies): a reply that is only "cancel" or "abort" is the person's word, kept and read like any other (#2370)
  • fix(install): a team's AGENTS.md that names AI-DLC is kept, and AI-DLC's block is added after it (#2373)
  • fix(sessions): when Claude Code, Codex, Kiro CLI or a hook names the chat, the engine no longer walks the process chain it would ignore (#2371)
  • fix(engine): a scope switch on finished work starts the stage it adds, and its gate opens (#2261) (#2369)
  • fix(source): a cache the project does not ignore can be left out after the refusal, and the refusal says how (#2244) (#2367)
  • fix(statusline): AI-DLC's status line is your choice; setup asks once and nothing replaces yours unasked (#1407) (#2366)
  • fix(sessions): on macOS the engine finds your chat with one process lookup per command, so a busy Mac keeps your words and your settings with the right chat (#2368)
  • fix(engine): the agent proposes the plan: the keyword classifier leaves the engine and the plan question names the plan the agent read (#2359)
  • fix(state): get --optional reads a field that may be absent without logging an error (#2278)
  • fix(aida): judge by the product picture and the tenets; review on open and on request (#2267)
  • fix(ci): the nightly no longer selects the two on-demand corpus checks, so five claude-sdk jobs stop failing on files that cannot run there (#2364)
  • test(live): t29 finds the error in any key order, and t176 answers the new-work question with the separate-work option wherever the agent puts it (#2365)
  • fix(state): a read of a field the work does not set leaves no error in the audit record (#2358)
  • fix(config): a Kiro CLI that both Kiro rows detect defaults to the row that pins the v3 engine (#2362)
  • fix(sensor): without User Stories a Unit traces its own story-map rows, not every FR/NFR (#2282) (#2363)
  • fix(opencode): setup and doctor say once when a global short-form permission rule is replaced in the project (#2357)
  • fix(kiro): recommended setup keeps the Kiro effort you saved, and Kiro IDE's row sets Kiro CLI's session model (#2268)
  • fix(kiro-ide): the Kiro CLI hooks-off step tells the person to type /aidlc after the restart (#2259)
  • fix(plugins): a plugin stage that names a core agent composes on Kiro IDE (#2325)
  • fix(engine): a fresh first turn is not read as hooks-off before hooks had a chance to run (#2354)

Source commit: 82a8448

Nightly test report

Full Suite failed for 82a8448ae2ed in run 38310246031. The preview build does not wait for these tests.

Failed legs: live_linux (failure), live_macos (failure), live_windows (failure).

Failing tests

  • t-e2e-isolated-runner (1 failed assertions) in live-claude-tui-1-Windows, live-claude-tui-2-Linux, live-claude-tui-2-macOS
  • t-tui-compose-front.serial (1 failed assertions) in live-claude-tui-1-Windows, live-claude-tui-2-Linux, live-claude-tui-3-macOS
  • t-tui-custom-harness.serial (1 failed assertions) in live-claude-tui-1-Linux, live-claude-tui-2-Windows, live-claude-tui-3-macOS
  • t-tui-journey-orientation-windows.serial (1 failed assertions) in live-claude-tui-2-Windows
    • t-tui-journey-orientation Windows startup > three native Windows launches stay within 20000ms and reap each process tree
  • t-tui-render-complete.serial (1 failed assertions) in live-claude-tui-3-macOS
  • t-tui-t101-memory-lifecycle.serial (1 failed assertions) in live-claude-tui-1-Linux, live-claude-tui-3-Windows, live-claude-tui-3-macOS
  • t-tui-t139-revision-loop-idempotency.serial (1 failed assertions) in live-claude-tui-1-Linux, live-claude-tui-1-Windows, live-claude-tui-1-macOS
  • t-tui-t24-stage-jump.serial (1 failed assertions) in live-claude-tui-2-macOS, live-claude-tui-3-Linux, live-claude-tui-3-Windows
  • t-tui-t27-depth-override.serial (1 failed assertions) in live-claude-tui-2-Windows, live-claude-tui-3-Linux, live-claude-tui-3-macOS
  • t-tui-t29-env-scope.serial (1 failed assertions) in live-claude-tui-1-macOS, live-claude-tui-2-Linux, live-claude-tui-3-Windows
  • t-tui-t50-bugfix-scope.serial (1 failed assertions) in live-claude-tui-2-macOS, live-claude-tui-3-Linux, live-claude-tui-3-Windows
  • t-tui-t51-poc-scope.serial (1 failed assertions) in live-claude-tui-2-Linux, live-claude-tui-2-Windows, live-claude-tui-3-macOS
  • t-tui-t58-workshop-scope.serial (1 failed assertions) in live-claude-tui-1-macOS, live-claude-tui-3-Linux, live-claude-tui-3-Windows
  • t-tui-t73-intent-capture.serial (1 failed assertions) in live-claude-tui-1-Linux, live-claude-tui-2-macOS, live-claude-tui-3-Windows
  • t-tui-t74-requirements-analysis.serial (1 failed assertions) in live-claude-tui-2-Windows, live-claude-tui-3-Linux, live-claude-tui-3-macOS
  • t-tui-windows-user-settings-isolation.serial (1 failed assertions) in live-claude-tui-1-Windows
  • t-tui-workshop.serial (1 failed assertions) in live-claude-tui-1-macOS, live-claude-tui-2-Linux, live-claude-tui-2-Windows
  • t138-scope-exclusion-counts (1 failed assertions) in live-claude-sdk-2-Windows
    • t138 scope-exclusion counts (metamorphic invariant, sdk) > every SKIP-for-security-patch stage emits zero STAGE_STARTED (SKIP set derived from scope-grid.json)
  • t141-enterprise-scope-routing (1 failed assertions) in live-claude-sdk-2-Windows, live-claude-sdk-2-macOS, live-claude-sdk-4-Linux
    • t141 enterprise scope routing (sdk live, MR9) > enterprise init records the scope and routes to the first post-init stage from scope-grid.json
  • t55-workflow-init-then-resume (1 failed assertions) in live-claude-sdk-1-Linux, live-claude-sdk-1-macOS, live-claude-sdk-2-Windows
    • t55 /aidlc creation (--scope bugfix) then resume continuity (sdk) > creation establishes state; a second scope turn resumes from it (init stages persist, audit grows across sessions)
  • t59-workflow-depth-override (1 failed assertions) in live-claude-sdk-2-Linux, live-claude-sdk-2-Windows, live-claude-sdk-2-macOS
    • t59 /aidlc --scope bugfix --depth comprehensive depth override (sdk) > depth comprehensive overrides the bugfix Minimal default at init, records bugfix scope + WORKFLOW_STARTED

Runner errors

  • live-claude-tui-1-Linux: INCOMPLETE: required selected coverage was not fully exercised
  • live-claude-tui-1-Windows: INCOMPLETE: required selected coverage was not fully exercised
  • live-claude-tui-1-macOS: INCOMPLETE: required selected coverage was not fully exercised
  • live-claude-tui-2-Linux: INCOMPLETE: required selected coverage was not fully exercised
  • live-claude-tui-2-Windows: INCOMPLETE: required selected coverage was not fully exercised
  • live-claude-tui-2-macOS: INCOMPLETE: required selected coverage was not fully exercised
  • live-claude-tui-3-Linux: INCOMPLETE: required selected coverage was not fully exercised
  • live-claude-tui-3-Windows: INCOMPLETE: required selected coverage was not fully exercised
  • live-claude-tui-3-macOS: INCOMPLETE: required selected coverage was not fully exercised

Failed jobs

Full logs are in the run 38310246031 full-suite-* artifacts.

Check out latest releases or
releases around awslabs/aidlc-workflows v2.11.1-preview.20261012.1

Don't miss a new aidlc-workflows release

NewReleases is sending notifications on new releases.

Get notifications