Devansh Jain
04e3d8739e
refactor: split session-context reminder into agent-info + session-context ( #1078 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-20 17:54:35 -08:00
github-actions[bot]
71d3298f69
chore: bump version to 0.16.2 [skip ci]
2026-02-21 01:01:35 +00:00
Kian Jones
c5033666b4
perf: eliminate redundant API calls on startup ( #1067 )
...
Co-authored-by: cpacker <packercharles@gmail.com >
Co-authored-by: Letta <noreply@letta.com >
2026-02-20 16:59:45 -08:00
Kian Jones
539ce175e5
feat: add --conv flag to listen subcommand ( #1070 )
2026-02-20 16:37:00 -08:00
Kian Jones
a16acfddeb
fix: memory palace fails if there's a $ in memory ( #1074 )
2026-02-20 16:31:41 -08:00
Charles Packer
150ff0d998
fix: no retry on quota limit errors ( #1072 )
2026-02-20 16:30:20 -08:00
Kevin Lin
2e8b00f8cc
fix: nesting level not correct in memory palace ( #1075 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-20 16:30:11 -08:00
Charles Packer
eca8459df4
fix: resolve env vars in reflection/history-analyzer commit trailers ( #1076 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-20 16:29:57 -08:00
Devansh Jain
6efcbd806f
fix: Disable all shared reminders for subagents ( #1073 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-20 16:24:36 -08:00
cthomas
1e024c34f9
fix: note single line vs multi line for git cred skill ( #1071 )
2026-02-20 16:13:51 -08:00
jnjpng
a7ec74612f
feat: log chunks for feedback ( #1068 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-20 16:04:38 -08:00
Kevin Lin
6b085ba791
fix: skip auto-open memory palace in tmux on macOS ( #1069 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-20 15:53:05 -08:00
Shubham Naik
4f8690f98d
fix: never auto-approve interactive tools in listen mode ( #1066 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-20 13:26:57 -08:00
Charles Packer
fa6ac1eb3a
fix: clear isExecutingTool on interrupt to unblock slash commands ( #1065 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-20 13:04:04 -08:00
github-actions[bot]
46246ebd39
chore: bump version to 0.16.1 [skip ci]
2026-02-20 20:59:07 +00:00
Charles Packer
a13fb37d87
feat: auto-enable memfs from server-side tag on new machines ( #1063 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-20 12:57:40 -08:00
jnjpng
4569382d20
fix(models): set max_output_tokens for GPT-5 reasoning variants ( #1064 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-20 12:55:59 -08:00
Shubham Naik
924ae8e8bc
feat: add approval loop and improved conversationId handling for list… ( #1057 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-20 12:30:20 -08:00
jnjpng
3fed452834
fix: map provider for generic LLM streaming errors ( #1062 )
2026-02-20 12:17:27 -08:00
Ari Webb
d9b35895ff
fix: zai error handling ( #1047 )
2026-02-20 12:01:43 -08:00
Charles Packer
b622eca198
feat(web): add Memory Palace static viewer ( #1061 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-20 12:00:55 -08:00
Kevin Lin
2da31bf2f7
fix: update /memory command to say "memory" instead of "memory blocks" ( #1058 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-20 11:02:40 -08:00
Charles Packer
66cd245156
fix: default headless mode to new conversation to prevent 409 race ( #1051 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-19 20:52:59 -08:00
Sarah Wooders
a1db9acbeb
docs(skills): prefer repo-local memfs auth and warn on global helper conflicts ( #1054 )
2026-02-19 20:52:12 -08:00
jnjpng
2dada876e9
fix: use getVersion() for feedback and telemetry version reporting ( #1048 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-19 16:56:59 -08:00
Kian Jones
49c96c3e46
chore: modify explore agent description for clarity ( #1046 )
2026-02-19 16:49:44 -08:00
Charles Packer
5378cc6dd8
fix: console error leak ( #1049 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-19 16:49:21 -08:00
Ari Webb
b7f381c183
fix: better error formatting for init ( #1044 )
2026-02-19 15:44:37 -08:00
Charles Packer
aae7da834e
revert: "revert: "fix: disable hidden SDK retries for streaming POSTs"" ( #1041 )
2026-02-19 12:59:54 -08:00
github-actions[bot]
b36b6293be
chore: bump version to 0.16.0 [skip ci]
2026-02-19 20:58:02 +00:00
Charles Packer
277ce3cd41
fix: recover default agent creation when base tools are missing ( #1039 )
2026-02-19 12:47:03 -08:00
Kevin Lin
a496204409
feat: add Gemini 3.1 Pro Preview model ( #1037 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-19 12:34:23 -08:00
Charles Packer
b956a00301
revert: "fix: disable hidden SDK retries for streaming POSTs" ( #1038 )
2026-02-19 12:33:45 -08:00
Charles Packer
fb5efd9809
fix: disable hidden SDK retries for streaming POSTs ( #981 )
2026-02-19 12:07:51 -08:00
cthomas
7fdd163665
fix: dont retry on agents limit exceeded ( #1035 )
2026-02-19 11:57:56 -08:00
Kevin Lin
9920b3a42d
feat: gate memfs on self-hosted behind LETTA_MEMFS_LOCAL env var ( #1036 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-19 11:49:02 -08:00
Charles Packer
3655429549
fix(prompt): improve subagent guidance for codex ( #1034 )
2026-02-19 00:48:46 -08:00
Charles Packer
093657b63c
fix: auto toolset detect chatgpt_oauth as codex ( #1033 )
2026-02-19 00:28:32 -08:00
Charles Packer
aed7e5d119
fix: allow plan-file apply_patch paths in plan mode ( #1032 )
2026-02-18 23:55:05 -08:00
Charles Packer
f92279c2a5
fix: compact plan and todo tool header args ( #1031 )
2026-02-18 23:31:59 -08:00
Charles Packer
366c1a39c7
fix: preserve footer reasoning tier across restart ( #1030 )
2026-02-18 23:13:02 -08:00
Charles Packer
47e81433ff
feat: align TaskOutput UX with Bash output ( #1029 )
2026-02-18 22:53:41 -08:00
paulbettner
674e8c12c2
fix(tui): show elapsed time for running shell tools ( #1026 )
2026-02-18 20:01:47 -08:00
Charles Packer
943c20a495
fix: permissions observability rollout ( #1028 )
2026-02-18 19:59:51 -08:00
Charles Packer
dc25ce5573
fix: harden permissions matching and alias caching ( #1027 )
2026-02-18 19:49:53 -08:00
Zerubbabel
92d2955035
fix: allow Gemini/GLM tools in plan mode ( #1025 )
2026-02-18 19:31:53 -08:00
cpacker
6b7c59b0be
fix(tui): populate reasoning_effort from updateArgs as fallback when API omits it
...
The Letta API _to_model_settings() for Anthropic was not including
effort in the GET response (backend bug), so agentState.model_settings.effort
and llm_config.reasoning_effort both came back null after a /model switch.
deriveReasoningEffort then had nothing to work with.
Client-side fix: after updateAgentLLMConfig returns, merge reasoning_effort
from model.updateArgs (/model switch) and desired.effort (Tab cycle flush)
into the llmConfig state we set. This populates the fallback path in
deriveReasoningEffort reliably regardless of what the API echoes back.
The actual root cause is fixed in letta-cloud: _to_model_settings() now
includes effort=self.effort for Anthropic models.
👾 Generated with [Letta Code](https://letta.com )
Co-Authored-By: Letta <noreply@letta.com >
2026-02-18 19:00:07 -08:00
Charles Packer
35291d9094
fix(tui): footer reasoning tier, toolset naming, and selector highlight ( #1024 )
...
Co-authored-by: Letta <noreply@letta.com >
2026-02-18 18:48:25 -08:00
paulbettner
b871ff793d
fix(permissions): avoid stale YOLO mode during approvals ( #1023 )
2026-02-18 18:41:58 -08:00
Charles Packer
a0a5bbfc72
feat: inject toolset swap and slash-command context reminders ( #1022 )
2026-02-18 18:15:38 -08:00