Commit Graph

51 Commits

Author SHA1 Message Date
Arvin A
55d8b59a48
feat(core): Stream tool calls and ship M3 fixtures from LLM eval wire server (no-changelog) (#30983) 2026-05-27 14:53:43 +01:00
José Braulio González Valido
76c432c53f
fix(ai-builder): Default Switch to case-insensitive in builder hints (#31044)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-27 07:42:54 +00:00
Albert Alises
8bb5db3bbd
feat(core): Add runtime skills to Instance AI builders (no-changelog) (#30838) 2026-05-27 06:54:40 +00:00
Albert Alises
959f8ca53c
refactor(core): Remove web researcher sub-agent (no-changelog) (#31141) 2026-05-26 17:25:50 +00:00
José Braulio González Valido
700b32237f
feat(ai-builder): Surface WHAT-dimension binary checks per built workflow (no-changelog) (#30932)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-26 12:18:52 +01:00
José Braulio González Valido
96a9521394
ci: Use PR head ref in eval experiment names (no-changelog) (#30898)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-22 11:06:50 +00:00
José Braulio González Valido
ada126d9b7
test(ai-builder): Validate user-proxy tool outputs against api-types schemas (no-changelog) (#30905)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-21 14:15:44 +00:00
José Braulio González Valido
81ea56fa6b
test(ai-builder): Add multi-turn capability for IAI evals (no-changelog) (#30586)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-21 13:03:35 +00:00
José Braulio González Valido
e9b1c7c48f
chore(ai-builder): Tag Daytona sandboxes with name prefix and labels (no-changelog) (#30697)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-21 12:43:36 +00:00
Bernhard Wittmann
374e7ed0b2
ci: Fail Instance AI discovery evals only on zero-pass scenarios (no-changelog) (#30816) 2026-05-21 06:44:18 +00:00
oleg
d7d2cc1442
feat(core): Add native agent substrate (no-changelog) (#30015)
Signed-off-by: Oleg Ivaniv <me@olegivaniv.com>
2026-05-19 13:48:45 +00:00
Albert Alises
17b64da0f0
fix: Add Switch fallback output guidance for workflow builder (#30449) 2026-05-15 07:33:25 +00:00
Albert Alises
f1fd79f830
fix(core): Avoid unnecessary planner credential prompts (#30451) 2026-05-14 15:36:11 +00:00
José Braulio González Valido
9fcd5c5864
feat(ai-builder): Tool-driven eval mock handler with per-API quirks registry (#30135)
Some checks are pending
CI: Master (Build, Test, Lint) / Build for Github Cache (push) Waiting to run
CI: Master (Build, Test, Lint) / Unit tests (22.x) (push) Waiting to run
CI: Master (Build, Test, Lint) / Unit tests (24.14.1) (push) Waiting to run
CI: Master (Build, Test, Lint) / Unit tests (25.x) (push) Waiting to run
CI: Master (Build, Test, Lint) / Lint (push) Waiting to run
CI: Master (Build, Test, Lint) / Performance (push) Waiting to run
CI: Master (Build, Test, Lint) / Notify Slack on failure (push) Blocked by required conditions
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-14 14:29:10 +00:00
Albert Alises
c3e39f8504
fix(ai-builder): Guide builder to prefer httpBearerAuth for Bearer flows (#30309) 2026-05-13 17:24:56 +00:00
Mutasem Aldmour
2fd54d8230
feat(core): Curate workflow examples for the builder sandbox (#30025)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-13 06:45:39 +00:00
Bernhard Wittmann
c9ecef2a36
test: Add discovery eval guardrails for browser/computer-use dispatch (no-changelog) (#30236) 2026-05-12 14:29:35 +00:00
Milorad FIlipović
54d62bb4a1
fix(core): Update instance-ai evaluator to include pinned subnodes and allow all mcp tools (#30292) 2026-05-12 09:13:01 +00:00
Bernhard Wittmann
b445221c6a
feat: Computer-use evaluation harness (no-changelog) (#29797)
Co-authored-by: Elias Meire <elias@meire.dev>
2026-05-12 08:36:12 +00:00
José Braulio González Valido
95cf41c37c
chore(core): Enable Daytona sandbox in Instance AI evals (no-changelog) (#29931)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-12 07:43:04 +00:00
Albert Alises
bb73952fcc
fix(core): Defer credential setup during workflow builds (#30181) 2026-05-11 15:46:44 +00:00
Mutasem Aldmour
0feec2fea6
fix(core): Make placeholder() return string (no-changelog) (#30100)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-11 11:32:35 +00:00
Mutasem Aldmour
d0367a00e8
chore: Align pairwise eval builder with production handover (#30019)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-11 11:00:37 +00:00
José Braulio González Valido
30d9a168bc
feat(ai-builder): Add --prebuilt-workflows flag for eval CLI (no-changelog) (#29830)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-07 11:47:29 +00:00
Mutasem Aldmour
1cb7c591b3
chore: Match production builder step cap in pairwise eval (#29977)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-07 09:53:36 +00:00
Bernhard Wittmann
68560fbb9a
refactor: Extract shared eval helpers (no-changelog) (#29800) 2026-05-07 08:05:01 +00:00
José Braulio González Valido
2164afc5df
chore(ai-builder): Improve eval comparison alert clarity (no-changelog) (#29929)
Some checks are pending
CI: Master (Build, Test, Lint) / Build for Github Cache (push) Waiting to run
CI: Master (Build, Test, Lint) / Unit tests (22.x) (push) Waiting to run
CI: Master (Build, Test, Lint) / Unit tests (24.14.1) (push) Waiting to run
CI: Master (Build, Test, Lint) / Unit tests (25.x) (push) Waiting to run
CI: Master (Build, Test, Lint) / Lint (push) Waiting to run
CI: Master (Build, Test, Lint) / Performance (push) Waiting to run
CI: Master (Build, Test, Lint) / Notify Slack on failure (push) Blocked by required conditions
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-06 21:20:49 +00:00
Albert Alises
869dc32c15
feat(ai-builder): Speeds up Instance AI eval by parallelizing iterations and trimming mock handler (no-changelog) (#29839) 2026-05-06 08:15:33 +00:00
José Braulio González Valido
bbe3e2d148
feat(ai-builder): Add per-PR eval regression detection vs LangSmith baseline (#29456)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-06 08:15:08 +00:00
Mutasem Aldmour
fdceec21b9
feat: Add pairwise workflow eval pipeline (#29123)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Co-authored-by: Jaakko Husso <jaakko@n8n.io>
2026-05-04 13:26:27 +00:00
Charlie Kolb
f775604c25
refactor: Split up instance-ai confirmation endpoint DTO by action (#29179) 2026-05-04 10:47:38 +00:00
Albert Alises
c28d501ba1
fix(ai-builder): Stop builder from adding auth to inbound trigger nodes by default (#29648) 2026-05-04 10:25:17 +00:00
Albert Alises
625ed5e95a
fix(core): Harden Set node workflow SDK contract (#29568) 2026-04-30 12:10:44 +00:00
José Braulio González Valido
e7f3e6f771
feat(ai-builder): Add three new workflow eval test cases (no-changelog) (#29351)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-30 08:11:04 +00:00
Albert Alises
139b803dae
fix: Use explicit node references for AI memory session keys (#29473) 2026-04-30 07:26:36 +00:00
José Braulio González Valido
4fd68bfc99
ci(ai-builder): Parallelize Instance AI eval CI across multiple n8n containers (#29545)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-30 07:22:28 +00:00
José Braulio González Valido
54d9286d92
fix(ai-builder): Filter LangSmith eval dataset by local file slugs (#29507)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-29 13:30:29 +00:00
José Braulio González Valido
e503587854
docs: Restructure instanceAI evals README for first-time users (no-changelog) (#29095)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-28 07:22:45 +00:00
José Braulio González Valido
29cdd011b0
fix(ai-builder): Honor --timeout-ms across the eval harness (no-changelog) (#29219)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-27 12:49:01 +00:00
José Braulio González Valido
250b718936
fix(ai-builder): Stop tombstoning example UUIDs in LangSmith dataset sync (no-changelog) (#29112)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-27 08:52:17 +00:00
José Braulio González Valido
540faa7c1d
fix(ai-builder): Update eval dataset prompts to match scenario checklists (no-changelog) (#29011)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-24 09:55:18 +00:00
Benjamin Schroth
c961849226
feat(ai-builder): Add sub-agent evaluation harness with binary checks (no-changelog) (#28289)
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-24 07:50:46 +00:00
Albert Alises
8b33424d0f
fix(ai-builder): Stop treating empty defaults as satisfying required for the Split node (#28978) 2026-04-23 16:22:19 +00:00
José Braulio González Valido
89e9117d39
fix(ai-builder): Expose outputCount when eval node output is truncated (no-changelog) (#28977)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-23 11:06:27 +00:00
José Braulio González Valido
16e5f9572f
feat(ai-builder): Add LangSmith integration for workflow eval tracking (no-changelog) (#28835)
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-23 08:47:02 +00:00
Luca Mattiazzi
714981eea3
feat: Add multiple runs to instanceAI eval (no-changelog) (#28493)
Co-authored-by: cubic-dev-ai[bot] <191113872+cubic-dev-ai[bot]@users.noreply.github.com>
2026-04-22 07:11:10 +00:00
José Braulio González Valido
560f300716
test: Add Instance AI workflow evals CI pipeline (no-changelog) (#28366)
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-20 14:15:41 +00:00
José Braulio González Valido
ac922fa38c
feat(ai-builder): Improve eval verifier and mock handler reliability (no-changelog) (#28255)
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-10 13:57:32 +00:00
José Braulio González Valido
91b01d27b9
feat(ai-builder): Fix IF/Switch/Filter node misconfiguration in builder (no-changelog) (#28172)
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-08 14:35:43 +00:00
José Braulio González Valido
fef91c97dd
feat(ai-builder): Add --keep-workflows flag and fix eval execution errors (no-changelog) (#28129)
Some checks are pending
Build: Benchmark Image / build (push) Waiting to run
CI: Master (Build, Test, Lint) / Build for Github Cache (push) Waiting to run
CI: Master (Build, Test, Lint) / Unit tests (22.x) (push) Waiting to run
CI: Master (Build, Test, Lint) / Unit tests (24.14.1) (push) Waiting to run
CI: Master (Build, Test, Lint) / Unit tests (25.x) (push) Waiting to run
CI: Master (Build, Test, Lint) / Lint (push) Waiting to run
CI: Master (Build, Test, Lint) / Performance (push) Waiting to run
CI: Master (Build, Test, Lint) / Notify Slack on failure (push) Blocked by required conditions
Util: Sync API Docs / sync-public-api (push) Waiting to run
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-07 17:35:04 +00:00