Repository navigation
test(agentcore): cover execution loop behavior - #100
Merged
Fodesu merged 3 commits intoOct 3, 2026
Merged
Conversation
Fodesu
force-pushed
the
feat/agent-runtime-loop-tests
branch
from
October 3, 2026 14:43
76b76db to
c252f1c
Compare
Fodesu
force-pushed
the
10000
feat/agent-runtime-loop-tests
branch
from
October 3, 2026 14:55
c252f1c to
b247cf3
Compare
Fodesu
force-pushed
the
feat/agent-runtime-loop-tests
branch
from
October 3, 2026 15:35
b247cf3 to
67392a7
Compare
Fodesu
force-pushed
the
feat/agent-runtime-loop-tests
branch
2 times, most recently
from
October 3, 2026 16:05
1586f1e to
570c1db
Compare
Fodesu
force-pushed
the
feat/agent-runtime-loop-tests
branch
from
October 3, 2026 16:37
570c1db to
fa82bc4
Compare
Fodesu
force-pushed
the
feat/agent-runtime-loop-tests
branch
from
October 3, 2026 16:54
fa82bc4 to
ae38c4a
Compare
Fodesu
force-pushed
the
feat/agent-runtime-loop-tests
branch
from
October 3, 2026 17:04
ae38c4a to
b3b3ec8
Compare
Fodesu
force-pushed
the
feat/agent-runtime-loop-tests
branch
from
October 3, 2026 17:11
b3b3ec8 to
3465e41
Compare
Fodesu
marked this pull request as ready for review
October 3, 2026 17:20
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds behavioral tests for the execution Loop and the owner handle. The suite exercises the public Loop contract across model planning, tool scheduling, effect settlement, observation, cancellation, ownership, and recovery-related edge cases.
Coverage
The loop tests verify that preparation freezes the request before dispatch, that start barriers precede external work, and that a retryable dispatch refusal is bounded by the configured policy. Model outcomes cover success, provider failure, malformed request/result handling, cancellation, missing frozen bodies, and unknown executor results. Tool scenarios cover sequential and bounded-parallel scheduling, validation before start, known versus unknown failures, target resolution, aliases, and stable CallID derivation.
Ownership tests check per-Run serialization, takeover behavior, stale outcomes, late outcomes after terminal state, and the distinction between a process that owns a dispatch and one that must leave recovery to the next owner. Progress and observation tests ensure provisional events carry effect generation/sequence while committed facts are emitted through the durable path. Compaction and in-turn cases exercise the
BeforePrepareseam without making the Loop own chatlog policy.Regression and feature tables encode the expected state-machine transitions and rejection behavior, including unknown models, unknown tools, approvals, cancellation, and tool-step retention. The owner handle tests cover the command-side ownership boundary used by the Loop.
Scope
This PR is test-only and adds no Loop production code.