ai-engineering-loop 1.13.0 → 1.14.0

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 (60) hide show
  1. package/.agents/workflows/ai-engineering-loop.md +47 -11
  2. package/.claude/commands/ai-engineering-loop.md +1 -1
  3. package/.claude/skills/ai-engineering-loop/SKILL.md +26 -44
  4. package/.claude/skills/ai-engineering-loop/references/loop-commands.md +20 -0
  5. package/.claude/skills/ai-engineering-loop/references/loop-deliver.md +8 -0
  6. package/.claude/skills/ai-engineering-loop/references/loop-goal.md +10 -0
  7. package/.claude/skills/ai-engineering-loop/references/loop-maker.md +8 -0
  8. package/.claude/skills/ai-engineering-loop/references/loop-review.md +18 -0
  9. package/.claude/skills/ai-engineering-loop/references/loop-risky.md +9 -0
  10. package/.claude/skills/ai-engineering-loop/references/loop-verify.md +20 -0
  11. package/.gemini/skills/ai-engineering-loop/SKILL.md +24 -37
  12. package/.gemini/skills/ai-engineering-loop/references/loop-commands.md +20 -0
  13. package/.gemini/skills/ai-engineering-loop/references/loop-deliver.md +8 -0
  14. package/.gemini/skills/ai-engineering-loop/references/loop-goal.md +10 -0
  15. package/.gemini/skills/ai-engineering-loop/references/loop-maker.md +8 -0
  16. package/.gemini/skills/ai-engineering-loop/references/loop-review.md +18 -0
  17. package/.gemini/skills/ai-engineering-loop/references/loop-risky.md +9 -0
  18. package/.gemini/skills/ai-engineering-loop/references/loop-verify.md +20 -0
  19. package/.grok/skills/ai-engineering-loop/SKILL.md +29 -61
  20. package/.grok/skills/ai-engineering-loop/references/loop-commands.md +20 -0
  21. package/.grok/skills/ai-engineering-loop/references/loop-deliver.md +8 -0
  22. package/.grok/skills/ai-engineering-loop/references/loop-goal.md +10 -0
  23. package/.grok/skills/ai-engineering-loop/references/loop-maker.md +8 -0
  24. package/.grok/skills/ai-engineering-loop/references/loop-review.md +18 -0
  25. package/.grok/skills/ai-engineering-loop/references/loop-risky.md +9 -0
  26. package/.grok/skills/ai-engineering-loop/references/loop-verify.md +20 -0
  27. package/CHANGELOG.md +69 -0
  28. package/README.md +40 -1
  29. package/README.npm.md +6 -2
  30. package/THIRD_PARTY_NOTICES.md +29 -0
  31. package/adapters/dot/skills/dot-dev-skill-router/SKILL.md +24 -16
  32. package/adapters/project-onboarding/SKILL.md +27 -0
  33. package/adapters/project-onboarding/references/output.md +29 -0
  34. package/adapters/quality/skills/ael-code-simplifier/SKILL.md +43 -0
  35. package/adapters/quality/skills/ael-receiving-code-review/SKILL.md +37 -0
  36. package/bin/ai-engineering-loop.js +979 -69
  37. package/core/root-cause-analysis.md +6 -0
  38. package/core/runtime-safety.md +30 -3
  39. package/evals/cases/21-tier-t0-docs.json +34 -0
  40. package/evals/cases/22-tier-t1-small-pass.json +30 -0
  41. package/evals/cases/23-tier-t1-blocker.json +26 -0
  42. package/evals/cases/24-tier-t1-auth-escalates.json +30 -0
  43. package/lib/auto-update.js +486 -0
  44. package/lib/budget.js +31 -1
  45. package/lib/cli-args.js +84 -0
  46. package/lib/docs.js +91 -0
  47. package/lib/doctor.js +7 -1
  48. package/lib/gates.js +114 -1
  49. package/lib/host-skill-build.js +121 -0
  50. package/lib/migrations.js +84 -0
  51. package/lib/project-registry.js +148 -0
  52. package/lib/run-state.js +6 -3
  53. package/lib/sync-hosts.js +325 -42
  54. package/lib/tier.js +134 -0
  55. package/lib/update-manager.js +17 -1
  56. package/lib/update-notice.js +192 -0
  57. package/lib/user-state.js +80 -0
  58. package/lib/verification-recorder.js +44 -0
  59. package/migrations/README.md +21 -0
  60. package/package.json +7 -2
@@ -0,0 +1,43 @@
1
+ ---
2
+ name: ael-code-simplifier
3
+ description: Use when the user explicitly asks to simplify, tidy, or de-slop a finished diff without changing behavior. Does NOT review for defects (use the Devil's Advocate), refactor architecture, rename public APIs, or run automatically after edits.
4
+ ---
5
+
6
+ # AEL Code Simplifier
7
+
8
+ Rewrite *how* changed code reads; never *what* it does. Behavior, contracts, status codes, queries, and public names stay identical.
9
+
10
+ Optional in the AI Engineering Loop: only at T1/T2, only on request, after Maker and **before** `verify`, so verification and the Devil's Advocate see the final diff.
11
+
12
+ ## Scope
13
+
14
+ 1. Only files in `git diff` (staged and unstaged) or paths the user named. Do not open neighbors "while you are there".
15
+ 2. Follow the target repository's own conventions (`.ai-engineering-loop/conventions.md`, `AGENTS.md`, `CLAUDE.md`), not personal taste.
16
+ 3. Keep architectural layers separate. Never merge or skip a layer.
17
+
18
+ ## Allowed changes
19
+
20
+ - Remove dead code, unused imports, unreachable branches, and debug output added by this diff.
21
+ - Flatten needless nesting with early returns; replace clever one-liners with plain code.
22
+ - Collapse duplicated logic introduced by this diff into one local helper.
23
+ - Rename local variables whose names hide intent.
24
+ - Delete comments that restate the code; keep comments that explain *why*.
25
+ - Replace hand-rolled code with an existing helper the repo already uses.
26
+
27
+ ## Not allowed
28
+
29
+ - Behavior or output changes, including error messages that callers or tests match.
30
+ - File moves, public API renames, new dependencies, new abstractions or frameworks.
31
+ - Changing or deleting tests to make them pass; new tests are Maker work.
32
+ - Formatting churn outside changed hunks.
33
+
34
+ If the code cannot get simpler without changing behavior, stop and say so; that is a separate task.
35
+
36
+ ## Procedure
37
+
38
+ 1. List the changed files and the hunks you will touch.
39
+ 2. Apply the smallest edits that make each hunk easier to read.
40
+ 3. Run the repository's fast checks (unit tests, typecheck, lint) and show the output. Any new failure means revert that edit.
41
+ 4. Report: files touched, what was simplified, and confirmation that behavior is unchanged with command evidence.
42
+
43
+ Inside a loop run, the simplified diff still goes through `verify`, claimed-vs-reality, and review; this skill never replaces a gate.
@@ -0,0 +1,37 @@
1
+ ---
2
+ name: ael-receiving-code-review
3
+ description: Use when acting on review feedback (Devil's Advocate findings during ITERATE, or human MR/PR comments) before changing code. Does NOT perform a review (use the Devil's Advocate) or decide PASS/ITERATE (use the Judge).
4
+ ---
5
+
6
+ # AEL Receiving Code Review
7
+
8
+ Feedback is a claim to check, not an order to obey. Verify each item against the code before you change anything.
9
+
10
+ ## For every finding or comment
11
+
12
+ 1. **Restate** it in one line with the exact file and line it refers to. If you cannot locate it, ask for the location; do not guess.
13
+ 2. **Verify** it against the current code, the frozen Goal Contract, and the command evidence:
14
+ - reproduce it (a failing test, a command, or a concrete input), or
15
+ - show why it does not apply (the code path, a passing test, or the AC row that excludes it).
16
+ 3. **Classify** it as one of:
17
+ - `valid`: fix it test-first (red test, then fix).
18
+ - `invalid`: decline with evidence.
19
+ - `out-of-scope`: valid but outside the frozen Goal; record it in the handoff or as a follow-up, and do not widen the diff.
20
+ - `unclear`: ask one precise question.
21
+ 4. **Respond** with that classification and evidence. No performative agreement ("Great catch!") and no silent fixes.
22
+
23
+ ## Rules
24
+
25
+ - Do not apply a suggestion you have not verified, even from a senior reviewer or a stronger model.
26
+ - Do not argue from preference. Disagreement needs evidence: code, tests, specs, or measurements.
27
+ - Fix one finding at a time and re-run the relevant check after each, so a regression is attributable.
28
+ - A finding that contradicts the frozen Goal Contract goes back to the human; never edit the contract yourself.
29
+ - In the AI Engineering Loop, every fixed finding is re-verified with `verify` and reviewed again by a **new** Devil's Advocate. Iteration 3 is the ceiling.
30
+
31
+ ## Reply template (MR/PR)
32
+
33
+ ```
34
+ F-<id> <valid|invalid|out-of-scope|unclear>: <one-line reason>
35
+ Evidence: <test name / command + exit code / file:line>
36
+ Change: <commit or "none">
37
+ ```