Repository navigation
chore(vendor): pin tinyagents to main - #8330
Conversation
Moves vendor/tinyagents from ed2880b (sanil-23's pin on epic/8089-test-ownership) to tinyagents main 1f24dbe, the merge of tinyhumansai/tinyagents#570, which brought that epic into main. A fast-forward of 94 commits. The nested pins move with it (tinyinference f5c0e56, tinytools 75e42ba). Lockfiles follow in their own commit.
|
Note Currently processing new changes in this PR. This may take a few minutes, please wait... ⚙️ Run configuration
⛔ Files ignored due to path filters: 1
📒 Files selected for processing: 3
Comment |
`cargo metadata` adds `memoffset` to nix 0.29.0's dependency list in crates/openhuman-app/Cargo.lock. This is drift already on main (the same regeneration on main 6512f93 produces the identical line), not something the tinyagents move causes: no vendor manifest changed in that range, and the root lock is unchanged. tokio stays at 1.54.1.
tinyinference's ChatResponse gained a public `finish_reason` field (tinyinference#136), which reaches openhuman with this tinyagents pin. The seven struct literals in the agent tests now set it to `None`, which is what providers report when they give no stop reason. None of these tests asserts on the stop reason.
Tiny Sweeper reviewReviewed head: Verdict: Ready for maintainer review · Priority: none Diagnostics and supporting detailsFull summaryPull request purpose (title): chore(vendor): pin tinyagents to main. This pull request repins `vendor/tinyagents` from `ed2880b` to tinyagents `main` at `1f24dbe`, a 94-commit fast-forward that brings the wait tools (`WaitTool`, `WaitLoopTool`) and the agent_loop truncated-empty cache bypass into main; nested pins move tinyinference `e5d7092` → `f5c0e56` and tinytools `3a01641` → `75e42ba`, with `crates/openhuman-app/Cargo.lock` picking up one `memoffset` line. Because tinyinference's `ChatResponse` gained a public `finish_reason` field, the visible text hunks add `finish_reason: None` to the affected test struct literals in the two agent test files. No test execution is claimed. The latest-push delta is unavailable. Unreviewed paths
Review snapshot
Completeness: Complete What changedRepins the vendored tinyagents submodule to tinyagents `main` (`1f24dbe`), moving the nested tinyinference and tinytools pins forward and adding a single `memoffset` line to `crates/openhuman-app/Cargo.lock`. To keep the agent test fixtures compiling against tinyinference's new public `finish_reason` field, seven `ChatResponse` struct literals in `crates/openhuman-core/src/agent/agent_tests.rs` and `crates/openhuman-core/src/agent/agent_turn_loop_tests.rs` now set `finish_reason: None`. No production logic or assertion changes in the visible hunks. FeaturesNone identified with supported citations. TestsNo supported feature-to-test mapping was produced. Test execution is not inferred. FindingsNo active actionable findings. Before mergeNone. How this fits togetherflowchart LR
n0["ScriptedProvider<br/>changed"]:::changed
n1["text_response<br/>changed"]:::changed
n2["tool_response<br/>changed"]:::changed
n3["xml_tool_response<br/>changed"]:::changed
n4["..._fallback_tool_calls_and_persists_history<br/>changed"]:::changed
n5["turn_errors_on_empty_or_null_text_response<br/>changed"]:::changed
n6["turn_preserves_text_alongside_tool_calls<br/>changed"]:::changed
n7["build_agent_with"]:::impacted
n8["build_agent_with_config"]:::impacted
n9["ChatResponse"]:::impacted
n10["history_is_not_trimmed_by_message_count"]:::impacted
n11["turn_emits_checkpoint_at_max_iterations"]:::impacted
n0 -->|uses| n9
n1 -->|uses| n9
n2 -->|uses| n9
n3 -->|uses| n9
n4 -->|calls| n7
n4 -->|tests| n7
n5 -->|calls| n7
n5 -->|tests| n7
n6 -->|calls| n7
n6 -->|tests| n7
n10 -->|calls| n8
n10 -->|tests| n8
n11 -->|calls| n8
n11 -->|tests| n8
classDef changed fill:#0d4429,stroke:#238636,color:#e6edf3
classDef impacted fill:#161b22,stroke:#6e7681,color:#c9d1d9
classDef flagged fill:#5a1e02,stroke:#d93f0b,color:#ffffff
classDef blocking fill:#67060c,stroke:#f85149,color:#ffffff
Agent review detailscritique
security
tests
commits
description
Evidence and run details
|
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
![tinysweeper[bot]](https://avatars.githubusercontent.com/in/4526585?s=60&v=4){kind=small}
tinysweeper
Bot
added
the
priority: p3
Summary
Moves
vendor/tinyagentsfromed2880bto tinyagentsmain(1f24dbe), so that openhuman builds tinyagents frommaininstead of a fork branch.ed2880bis sanil-23's pin commit on top ofepic/8089-test-ownership. Merge epic/8089-test-ownership into main tinyagents#570 merged that whole epic into tinyagentsmain, including the wait tools openhuman imports (WaitTool,WaitLoopTool) and the agent_loop truncated-empty cache bypass.1f24dbeis that merge.ed2880bis an ancestor of1f24dbe, so this is a fast-forward of 94 commits and nothing openhuman runs today is dropped.e5d7092→f5c0e56, tinytools3a01641→75e42ba. tinyliveagents is unchanged.This follows #8325, which moved the other source-built vendor pins (TA-MAIN-2325).
Status
cargo metadataon the CI host. The root lock is unchanged.crates/openhuman-app/Cargo.lockgets one line,memoffsetinnix 0.29.0's dependencies. That line is drift already on main: the same regeneration on main6512f93produces it. tokio stays at 1.54.1.ChatResponsegained a publicfinish_reasonfield (tinyinference#136). The seven struct literals inagent_tests.rsandagent_turn_loop_tests.rsnow set it toNone.23af4a1, merged onto main418c50a, against the6512f93baseline: no new failures.clippy-product(openhuman-core, product features,-D warnings) passes, so thetinyagents_harness_tools::{WaitLoopTool, WaitTool}imports compile. The already-red coverage and gate-contract lanes fail with the same signatures as main.Tests
Checked on the shared CI host against a main baseline. Only failures that are new against main are attributed to this PR.
Summary by CodeRabbit