Repository navigation
ci: add PyPI release workflow for v0.1.0 [OMN-7647] - #8
Merged
Merged
Conversation
Add release.yml (tag-triggered build + publish via uv) and release-dry-run.yml matching org-wide pattern from omnibase_core and omnimemory. Requires PYPI_TOKEN secret to be configured.
|
Warning Rate limit exceeded
Your organization is not enrolled in usage-based pricing. Contact your admin to enable usage-based pricing to continue reviews beyond the rate limit, or try again in 3 minutes and 31 seconds. ⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (2)
📝 WalkthroughWalkthroughThe PR adds two GitHub Actions workflows for managing Python package releases. A dry-run workflow tests building distribution artifacts without publishing, while a production workflow builds, publishes artifacts to PyPI, generates checksums, and creates GitHub releases with automated release notes. Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~12 minutes Poem
✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In @.github/workflows/release-dry-run.yml:
- Around line 8-11: The workflow marks the inputs.tag as required but never uses
it; either make it optional or actually consume it. Update the checkout step to
use inputs.tag as the ref (e.g., ref: ${{ inputs.tag }}) and add an early
validation step (run: ...) that reads pyproject.toml, extracts the version, and
compares it to inputs.tag (fail the job if they differ) so the dry-run truly
validates the specified tag before the build step.
In @.github/workflows/release.yml:
- Around line 85-92: The release step "Set release tag" (id: tag) currently
writes the resolved tag to the output variable "tag" but downstream logic
derives "prerelease" from github.ref_name; change that to derive prerelease from
the resolved output "tag" instead. Modify the "Set release tag" step to also
compute and export a "prerelease" output by inspecting the resolved tag value
(the same value echoed to $GITHUB_OUTPUT as "tag") and checking for the
prerelease identifier (e.g., contains "rc" or other prerelease suffix), so that
when github.event_name is "workflow_dispatch" the user-provided inputs.tag is
used to determine prerelease status rather than github.ref_name.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Lite
Run ID: d7052f8c-aec1-4172-886d-869b13d5a924
📒 Files selected for processing (2)
.github/workflows/release-dry-run.yml.github/workflows/release.yml
- release-dry-run.yml: add tag validation step so the required tag input is actually consumed and verified against pyproject.toml - release.yml: derive prerelease from steps.tag.outputs.tag instead of github.ref_name so manual dispatch with rc tags works correctly
jonahgabriel
enabled auto-merge
April 6, 2026 23:13
jonahgabriel
added this pull request to the merge queue
Apr 7, 2026
jonahgabriel
added a commit
that referenced
this pull request
Apr 21, 2026
… Pick #8] - Wraps /onex:agent_healthcheck skill as proper ONEX node - Polls TaskList, detects stalls, redispatches v2 agents - Writes friction events on escalation to Blocked - Entry point: omnimarket.nodes.node_worker_stall_recovery - Unit tests + documentation included
jonahgabriel
added a commit
that referenced
this pull request
Apr 22, 2026
… Pick #8] - Wraps /onex:agent_healthcheck skill as proper ONEX node - Polls TaskList, detects stalls, redispatches v2 agents - Writes friction events on escalation to Blocked - Entry point: omnimarket.nodes.node_worker_stall_recovery - Unit tests + documentation included
jonahgabriel
added a commit
that referenced
this pull request
Apr 22, 2026
… Pick #8] - Wraps /onex:agent_healthcheck skill as proper ONEX node - Polls TaskList, detects stalls, redispatches v2 agents - Writes friction events on escalation to Blocked - Entry point: omnimarket.nodes.node_worker_stall_recovery - Unit tests + documentation included
github-actions Bot
pushed a commit
to andywu42/omnimarket
that referenced
this pull request
Apr 23, 2026
…OMN-9404 Pick OmniNode-ai#4] (OmniNode-ai#371) * feat(node_kafka_topic_emit_probe): hourly Kafka topic health probes [OMN-9404 Pick OmniNode-ai#4] - Emits synthetic events for declared Kafka topics to verify producers/consumers - Catches silent EMIT_FAILED failures proactively (1hr vs 6 days reactive) - Entry point: omnimarket.nodes.node_kafka_topic_emit_probe - Uses omnibase_core event bus protocols - Unit tests + documentation included * feat(node_merge_effect): simple git merge resolver for DIRTY PRs [OMN-9404 Pick OmniNode-ai#5] - Attempts git merge origin/main for conflict-free PRs - Escalates to LLM when merge is ambiguous - Mechanical automation based on OmniNode-ai#1354/OmniNode-ai#1356 pattern - Entry point: omnimarket.nodes.node_merge_effect - Unit tests + documentation included * feat(node_worker_stall_recovery): agent stall auto-recovery [OMN-9404 Pick OmniNode-ai#8] - Wraps /onex:agent_healthcheck skill as proper ONEX node - Polls TaskList, detects stalls, redispatches v2 agents - Writes friction events on escalation to Blocked - Entry point: omnimarket.nodes.node_worker_stall_recovery - Unit tests + documentation included * fix(node_kafka_topic_emit_probe): add missing Python models, fix mypy errors + CodeRabbit issues - Create ModelKafkaProbeRequest, ModelMergeCommand, ModelStallRecoveryCommand as Pydantic models (were YAML-only; handlers imported Python modules that didn't exist) - Remove non-existent omnimarket.adapters.event_bus imports from handler_kafka_probe - Remove ProtocolStateStore() instantiation (Protocol class, cannot be instantiated) - Fix publish() call signature: protocol takes ProtocolEventMessage, not (topic, dict) - Remove per-topic asyncio.sleep(interval) that caused 3600s stall in tests and exceeded 120s node timeout; probe_interval_seconds is a scheduling hint, not an intra-loop delay - Fix hardcoded /Users/jonah path in handler_stall_recovery (rule 6) - Fix test mock in test_handle_healthy_agent to use side_effect so git --version succeeds while grep returns non-zero (simulating no dispatch log match) - Fix docs: correct test path in kafka-probe tracking doc, add blank lines around MD table - Replace asyncio.get_event_loop().time() with time.time() for wall-clock timestamps * ci: retrigger PR checks after rebase * style: normalize docstring whitespace in handler_kafka_probe * chore: trigger CI re-run [no-op] * fix(cr-threads): address 18 CodeRabbit issues — stall recovery correctness + probe metadata - Fix side_effect_class: read_only → write (node emits Kafka events) - Fix _check_stall: grep -l multi-file handling, timezone-aware comparison, context_threshold_pct now wired to stall trigger logic - Fix lookup failures returned as healthy → now return status: unknown - Fix dry_run stored on self → request-scoped local variable - Fix checkpoint path .yaml extension → .json (matches JSON payload) - Fix _redispatch_agent: attempt-indexed suffix prevents -v2 collision - Fix hardcoded /Users/jonah paths → _resolve_onex_state() via OMNI_HOME env - Add field bounds to model_stall_recovery_command (ge/le constraints) - Fix docs: idempotency claim (not idempotent) + MD058 blank lines around tables - Fix test: remove no-op `or True` assertion; add real healthy/unknown tests * fix(cr-threads): resolve all 9 unresolved CodeRabbit threads + Runtime Sweep Critical: - handler_kafka_probe: replace stubbed _emit_probe/_verify_consumer with real aiokafka producer/admin-client implementation; apply probe_interval_seconds as inter-topic delay - handler_merge: make dry_run request-scoped (was shared mutable instance state, race-prone); add TimeoutExpired handling and timeout= to all subprocess.run calls; abort merge state after dry-run and before no-conflict early-return Major: - handler_stall_recovery: verify grep (not git) in initialize() via shutil.which; defer mkdir out of _get_checkpoint_path into _save_checkpoint writer path; map 'unknown' status -> 'healthy' to stay within declared contract outputs - test_node_merge_effect: mock subprocess.run in test_handle_nonexistent_repo - test_node_worker_stall_recovery: update test to expect 'healthy' for lookup failures (contract-aligned) Runtime Sweep: - pyproject.toml: remove stale node_merge_sweep entry point (module was renamed to node_merge_sweep_compute in OmniNode-ai#373; leftover caused 1/129 broken entry points) * fix(ci): resolve 3 CI failures — stub __init__, aiokafka typing, kafka probe test mock - Remove empty __init__ from HandlerMergeEffect (lint stub-only check) - Add aiokafka to mypy overrides to silence import-untyped errors - Mock _emit_probe in unit tests so they don't require a live Kafka broker * fix(typecheck): remove stale type: ignore[import-untyped] for aiokafka — now covered by mypy override * fix(cr-threads): load default topics from contract.yaml, clarify probe_interval semantics Resolves 2 unresolved CodeRabbit threads on handler_kafka_probe.py: - Major: hardcoded topics in _all_declared_topics() — replaced with _load_default_topics() that reads default_topics from contract.yaml inputs section - Minor: probe_interval_seconds timeout concern — clarified in docstring and contract description that this is scheduling metadata, not execution time * style: ruff format handler_kafka_probe.py
This was referenced Jul 5, 2026
Merged
andywu42
pushed a commit
to andywu42/omnimarket
that referenced
this pull request
Jul 6, 2026
…infra venv (OmniNode-ai#1625) _resolve_skill_roots derived the default skill roots from _REPO_ROOT (Path(__file__).parents[5]). Installed into the canonical infra venv, _REPO_ROOT is the site-packages install dir, so the default pointed at <venv>/.../plugins/onex/skills — which never exists — and the audit hard-failed with SkillFunctionalAuditNoSkillsDiscoveredError. The 108 real skills live in the sibling omniclaude/plugins/onex/skills clone. Mirror _resolve_nodes_roots: resolve the skill libraries via the $OMNI_HOME sibling-repo layout, so resolution works even when _REPO_ROOT is a site-packages dir. Keep canonical-clone-relative fallbacks. Fail-fast preserved (CLAUDE.md rule OmniNode-ai#8): no silent wrong default. When $OMNI_HOME is unset AND no repo-relative path resolves, only nonexistent candidates remain, so _audit_skills raises the zero-skills hard fail rather than reporting a vacuous pass. _resolve_omni_home returns None (not a guessed path) when neither the env var nor a validated canonical-clone anchor is available. Tests: site-packages _REPO_ROOT + $OMNI_HOME resolves integration_sweep and dod_sweep; unset $OMNI_HOME + no path still raises; canonical-clone fallback resolves without the env var. Closes OMN-13993
andywu42
pushed a commit
to andywu42/omnimarket
that referenced
this pull request
Jul 6, 2026
…ion_sweep/dod_sweep (OmniNode-ai#1624) Both sweep orchestrators resolve their contracts/artifact root from ONEX_CC_REPO_PATH via _resolve_root. A stale/container value (the in-container mount /onex_change_control) leaks into local infra-venv runs where it has no contracts/ dir, so _resolve_root raised and the sweep never ran. Fall back to the canonical registry clone at $OMNI_HOME/onex_change_control when the env path lacks contracts/, before raising. Fail-fast (CLAUDE.md rule OmniNode-ai#8) is preserved: os.environ["OMNI_HOME"] raises KeyError when unset (never a silent default), and a fallback that itself lacks contracts/ still raises RuntimeError. The 'ONEX_CC_REPO_PATH not set at all' branch is unchanged. Parametrized tests cover both handlers: OMNI_HOME registry fallback (behavioral) + unset-OMNI_HOME fail-fast.
This was referenced Jul 16, 2026
jonahgabriel
added a commit
that referenced
this pull request
Jul 18, 2026
…1805) * feat(OMN-14462): add scripts/merge-proof local-proof env wrapper (F-20) Local proof gates (topic-naming lint) resolve their validator from a sibling omnibase_infra checkout. Under the mandated worktree layout the naive ../omnibase_infra sibling path does not exist, so a correct change fails locally until the hidden environment is reconstructed by hand — the natural workaround pressure being --no-verify, which rule #10 bans. scripts/merge-proof resolves OMNI_HOME + OMNIBASE_INFRA_PATH deterministically from its own on-disk location (rule #8) and, when it cannot, prints the exact export lines to run rather than proceeding on a silent wrong path. Modes: --check, --print-env, -- CMD, default (run topic-lint proof). Test: tests/scripts/test_merge_proof.py drives the real script via subprocess in an isolated tmp location (git env stripped per OMN-14746/14744). * test(OMN-14462): avoid literal path prefixes in merge-proof test * fix(OMN-14462): harden merge-proof exported environment * test(OMN-14462): make merge-proof topology hermetic --------- Co-authored-by: test <test@example.com>
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
1 participant
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
release.ymlworkflow: triggers onv*tag push or manual dispatch, builds withuv build, publishes to PyPI via token, creates GitHub Release with checksumsrelease-dry-run.ymlworkflow: manual dispatch to verify build without publishingPre-release checklist
PYPI_TOKENrepository secret on GitHub>=0.39.0, PyPI currently has 0.38.0)git tag v0.1.0 && git push origin v0.1.0Test plan
Summary by CodeRabbit