Cristina Poncela Cubeiro
5d6237ab34
fix: helper owns create
2026-07-27 17:58:34 +02:00
Cristina Poncela Cubeiro
3c2b71b570
refactor: search index swappable
2026-07-27 17:52:13 +02:00
Christian Klotz
a597371bda
Merge pull request #7117 from earendil-works/feat/coding-agent-evals
...
feat(coding-agent): add extension creation eval
2026-07-27 18:05:08 +03:00
Christian Klotz
b0864a66cd
fix(coding-agent): preserve eval run diagnostics
2026-07-27 18:00:53 +03:00
Christian Klotz
b56a35f90f
fix(coding-agent): project typed eval outputs
2026-07-27 17:40:14 +03:00
Cristina Poncela Cubeiro
99e34013d1
feat: auth print ( #7168 )
CI / build-check-test (push) Waiting to run
2026-07-27 16:13:06 +02:00
Cristina Poncela Cubeiro
0ee1ec0030
rename: may not be index
2026-07-27 16:12:43 +02:00
Christian Klotz
af19a9eea8
fix(coding-agent): simplify extension eval prompt
2026-07-27 15:30:37 +03:00
Christian Klotz
95d1c60fe2
fix(coding-agent): simplify multi-step eval input
2026-07-27 15:16:01 +03:00
Christian Klotz
9423204c6b
fix(coding-agent): clarify eval harness errors
2026-07-27 15:02:23 +03:00
Christian Klotz
8a97632459
fix(coding-agent): keep eval scenarios in test cases
2026-07-27 14:50:51 +03:00
Michael Yu
2efa728d2e
fix(coding-agent): support concurrent user bash cancellation ( #7103 )
2026-07-27 12:00:24 +02:00
Michael Yu
c2275d6703
fix(coding-agent): prevent duplicate messages on startup session switch ( #7110 )
2026-07-27 11:59:27 +02:00
Rafael
61da9e2f37
feat(ai): add manual redirect URL fallback to OpenRouter OAuth login ( #7114 )
...
The OpenRouter PKCE flow only completed through the loopback callback
server, so remote/headless sessions (e.g. pi over SSH) could never
finish login: the browser runs on another machine and cannot reach
127.0.0.1 on the host running pi.
Race the callback server against a manual_code prompt, matching the
existing Anthropic and OpenAI Codex flows. Users can paste the final
redirect URL (or the bare authorization code) and the flow exchanges
it with the same PKCE verifier. A claimed callback keeps priority so
a pasted code cannot interrupt an in-flight exchange.
The callback server object follows the OAuthServerInfo shape from the
OpenAI Codex flow: waitForCredential() resolves null when cancelWait()
hands the login over to manual entry, and close() is pure cleanup that
never settles the wait.
Refs #7078
2026-07-27 11:58:04 +02:00
Lucas Meijer
f9a49869c2
feat(ai): expose pending stop reason while streaming ( #7151 )
...
* feat(ai): expose pending streaming stop reason
Closes #7142
* manual tidy up of human facing copy
2026-07-27 11:56:15 +02:00
Jack Rose
60f6a8034d
feat(ai): add GitHub Copilot Claude Opus 5 support ( #7158 )
...
Route claude-opus-5 through the Anthropic Messages API with adaptive
thinking, the Copilot minimal thinking-level override, and pin the
extended 1M context window alongside the other Copilot Opus models.
2026-07-27 11:52:19 +02:00
Thomas Mustier
f08f58f5f9
fix(coding-agent): run coding-agent tests offline by default ( #7031 )
2026-07-27 11:27:43 +02:00
github-actions[bot]
5f9d025e7a
chore: approve contributors from issue #7143
2026-07-27 08:52:22 +00:00
github-actions[bot]
3cd39163cb
chore: approve contributors from issue #7132
2026-07-27 08:44:33 +00:00
Cristina Poncela Cubeiro
bd8dc676bf
Merge branch 'main' into feat/search
2026-07-27 10:05:39 +02:00
Cristina Poncela Cubeiro
b92e5e8614
feat: search index factory
2026-07-27 10:02:25 +02:00
Cristina Poncela Cubeiro
0f5a6e422b
feat: search index sqlite
2026-07-27 09:47:22 +02:00
Armin Ronacher
cee5ff7520
ref: remove openclaw reference from readme
2026-07-26 18:29:38 +02:00
Christian Klotz
a0ac81c0a1
fix(coding-agent): distinguish eval invariants
2026-07-26 00:02:34 +03:00
Christian Klotz
4b8accab8b
fix(coding-agent): simplify eval normalization
2026-07-25 23:47:07 +03:00
Christian Klotz
73c1696d95
fix(coding-agent): use focused eval harness
2026-07-25 23:34:20 +03:00
Christian Klotz
66fbdb1c59
feat(coding-agent): add extension creation eval
2026-07-25 23:09:22 +03:00
Armin Ronacher
5bc1c2c0a6
Add [Unreleased] section for next cycle
CI / build-check-test (push) Has been cancelled
2026-07-25 14:37:15 +02:00
Armin Ronacher
b4f293684b
Release v0.82.1
2026-07-25 14:37:11 +02:00
Armin Ronacher
6173017a7b
fix: drop harness-pi-ai and its legacy registry aliases from evals
...
The harness declared peer deps on the pre-rename @mariozechner/pi-ai and
@mariozechner/pi-agent-core names, so npm auto-installed our own code from the
registry under an alias. sync-versions bumped those aliases in lockstep, which
made every release fail because the new version is not published yet.
Build the eval harness on createHarness from vitest-evals core instead. The
pi-specific code was already ours; the harness only contributed agent-lifecycle
plumbing and unused tool replay.
2026-07-25 14:36:32 +02:00
Armin Ronacher
62a7074188
docs: audit changelogs for commits since v0.82.0
...
Add missing entries for the llama.cpp catalog cache and resource loader
directory fix, duplicate ai-side changes into coding-agent, add missing
external PR attributions, restore the 0.82.0 section that was modified
after release, and add the New Features section.
2026-07-25 14:09:45 +02:00
Duncan Ogilvie
58c0bc2fba
fix(coding-agent): exclude directories from resource loader ( #7106 )
...
This fixes the following warning:
Warning: Could not read /Users/duncan/Projects/AGENTS.md: Error: EISDIR: illegal operation on a directory, read
2026-07-25 14:05:16 +02:00
Armin Ronacher
b711e26616
fix: stop issue analysis from reporting failed runs as finished
...
Only comment the analysis summary when the pi run succeeded, add an explicit
failure comment, skip unchanged auth secret writes, and document that
PI_AUTH_UPDATE_TOKEN needs the Environments permission.
2026-07-25 12:41:22 +02:00
Armin Ronacher
4cf0a729c4
fix(ai): keep the underlying cause in ModelsError messages
...
Auth failures surfaced only the wrapper text, so CI logs showed a bare
"OAuth refresh failed for openai-codex" with no provider response.
2026-07-25 12:41:15 +02:00
Armin Ronacher
b1c444d956
feat(coding-agent): revalidate remote model catalogs with ETag
2026-07-25 12:26:19 +02:00
Alexey Zaytsev
518855dd50
feat(coding-agent): expose output padding to custom renderers ( #7045 )
...
Co-authored-by: Mario Zechner <badlogicgames@gmail.com>
2026-07-25 08:29:34 +02:00
David Brailovsky
2a2b0a39f0
fix(coding-agent): cache llama.cpp model catalog ( #7072 )
2026-07-25 08:27:05 +02:00
Burak Varlı
af3b934fa6
feat(ai): support Claude Opus 5 on Bedrock ( #7081 )
2026-07-25 08:25:23 +02:00
Christian Klotz
eafe11fb94
feat(coding-agent): add vitest eval harness ( #7085 )
2026-07-25 08:22:32 +02:00
github-actions[bot]
440689dfc7
chore: approve contributors from issue #7078
2026-07-25 06:15:33 +00:00
Christian Klotz
a3ee1d286e
fix(coding-agent): expose unavailable scoped models ( #7032 )
...
* fix(coding-agent): expose unavailable scoped models
Keep missing configured models editable in the scoped-model selector. This fixes #6949 .
* fix(coding-agent): reuse model scope diagnostics
* fix(coding-agent): classify model scope diagnostics
2026-07-25 07:19:56 +02:00
David Brailovsky
8eef62ed3e
fix tests
2026-07-24 19:05:08 +00:00
David Brailovsky
921c354352
add opus-5 models settings ( #7083 )
2026-07-24 20:57:46 +02:00
David Brailovsky
7304d8f8af
fix tests: use updated models
2026-07-24 18:52:56 +00:00
Armin Ronacher
a9f5b1c123
fix(ai): route Radius OAuth through gateway
CI / build-check-test (push) Waiting to run
2026-07-24 16:12:44 +02:00
Armin Ronacher
24e5cc04f4
fix(ai): support Anthropic bearer token env ( #6148 )
...
Fixes #5871
2026-07-24 15:13:22 +02:00
Armin Ronacher
7df73a00c6
Add [Unreleased] section for next cycle
2026-07-24 08:00:35 +02:00
Armin Ronacher
083e616212
Release v0.82.0
2026-07-24 08:00:31 +02:00
Armin Ronacher
e8d97fee1a
feat(ai): update generated image models
2026-07-24 07:59:40 +02:00
Armin Ronacher
e16172069f
docs(coding-agent): add model reload changelog entry
2026-07-24 07:56:05 +02:00
