-
Notifications
You must be signed in to change notification settings - Fork 2
chore: sync workflow templates #623
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Closed
Closed
Changes from all commits
Commits
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
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
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
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
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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -25,11 +25,13 @@ | |
| "agents-autofix-metrics", | ||
| "agents-verifier-metrics", | ||
| "agents-verifier-disposition-metrics", | ||
| "codex-cli-freshness", | ||
| } | ||
| _PREFIXED_ARTIFACT_FAMILIES = ( | ||
| "autopilot-metrics-", | ||
| "issue-optimizer-metrics-", | ||
| "issue-intake-format-metrics-", | ||
| "codex-cli-freshness-", | ||
| "verifier-terminal-disposition-", | ||
| "review-thread-terminal-disposition-", | ||
| ) | ||
|
|
@@ -323,6 +325,8 @@ def _classify_entry(entry: dict[str, Any]) -> str: | |
| return "terminal_disposition" | ||
| if schema == "workflows-verifier-followup-ledger/v1": | ||
| return "verifier_followup_ledger" | ||
| if schema == "workflows-codex-cli-freshness/v1": | ||
| return "codex_cli_freshness" | ||
| explicit = entry.get("metric_type") or entry.get("type") or entry.get("workflow") | ||
| if isinstance(explicit, str): | ||
| lowered = explicit.lower() | ||
|
|
@@ -506,6 +510,7 @@ def _summarise_verifier( | |
| terminal_sources = Counter() | ||
| verifier_models = Counter() | ||
| model_selection_reasons = Counter() | ||
| verifier_cli_versions = Counter() | ||
| unsupported_verifier_models = Counter() | ||
| unsupported_model_dispositions = Counter() | ||
| missing_verifier_model_metadata = Counter() | ||
|
|
@@ -574,6 +579,14 @@ def _summarise_verifier( | |
| ) | ||
| if model_selection_reason: | ||
| model_selection_reasons[str(model_selection_reason)] += 1 | ||
| cli_version = ( | ||
| entry.get("codex_cli_version") | ||
| or entry.get("llm_cli_version") | ||
| or entry.get("cli_version") | ||
| ) | ||
| cli_version_text = str(cli_version).strip() if cli_version is not None else "" | ||
| if cli_version_text: | ||
| verifier_cli_versions[cli_version_text.lower()] += 1 | ||
| verifier_mode = str(entry.get("verifier_mode") or "").strip().lower() | ||
| if verifier_mode: | ||
| verifier_modes[verifier_mode] += 1 | ||
|
|
@@ -623,6 +636,7 @@ def _summarise_verifier( | |
| "terminal_dispositions": terminal_dispositions, | ||
| "terminal_sources": terminal_sources, | ||
| "verifier_models": verifier_models, | ||
| "verifier_cli_versions": verifier_cli_versions, | ||
| "unsupported_verifier_models": unsupported_verifier_models, | ||
| "unsupported_model_dispositions": unsupported_model_dispositions, | ||
| "missing_verifier_model_metadata": missing_verifier_model_metadata, | ||
|
|
@@ -728,6 +742,54 @@ def _summarise_autopilot(entries: list[dict[str, Any]]) -> dict[str, Any]: | |
| } | ||
|
|
||
|
|
||
| def _summarise_codex_cli_freshness(entries: list[dict[str, Any]]) -> dict[str, Any]: | ||
| statuses = Counter() | ||
| packages = Counter() | ||
| pinned_versions = Counter() | ||
| latest_versions = Counter() | ||
| max_major_delta = 0 | ||
| max_minor_delta = 0 | ||
| max_patch_delta = 0 | ||
| update_targets = Counter() | ||
| for entry in entries: | ||
| status = str(entry.get("status") or "unknown") | ||
| statuses[status] += 1 | ||
| package = str(entry.get("package") or "unknown") | ||
| packages[package] += 1 | ||
| pinned = str(entry.get("pinned_version") or "unknown") | ||
| latest = str(entry.get("latest_version") or "unknown") | ||
| pinned_versions[pinned] += 1 | ||
|
Comment on lines
+755
to
+761
|
||
| latest_versions[latest] += 1 | ||
| delta = entry.get("version_delta") | ||
| if isinstance(delta, dict): | ||
| max_major_delta = max(max_major_delta, _safe_int(delta.get("major")) or 0) | ||
| max_minor_delta = max(max_minor_delta, _safe_int(delta.get("minor")) or 0) | ||
| max_patch_delta = max(max_patch_delta, _safe_int(delta.get("patch")) or 0) | ||
| targets = entry.get("update_targets") | ||
| if isinstance(targets, list): | ||
| for target in targets: | ||
| if not isinstance(target, dict): | ||
| continue | ||
| path = str(target.get("path") or "").strip() | ||
| if path: | ||
| update_targets[path] += 1 | ||
| return { | ||
| "records": len(entries), | ||
| "statuses": statuses, | ||
| "packages": packages, | ||
| "pinned_versions": pinned_versions, | ||
| "latest_versions": latest_versions, | ||
| "outdated_records": statuses.get("outdated", 0), | ||
| "latest_unavailable_records": statuses.get("latest-unavailable", 0), | ||
| "max_version_delta": { | ||
| "major": max_major_delta, | ||
| "minor": max_minor_delta, | ||
| "patch": max_patch_delta, | ||
| }, | ||
| "update_targets": update_targets, | ||
| } | ||
|
|
||
|
|
||
| def _format_counter(counter: Counter[str]) -> str: | ||
| if not counter: | ||
| return "n/a" | ||
|
|
@@ -765,6 +827,7 @@ def _bucket_entries(entries: list[dict[str, Any]]) -> dict[str, list[dict[str, A | |
| "verifier": [], | ||
| "terminal_disposition": [], | ||
| "verifier_followup_ledger": [], | ||
| "codex_cli_freshness": [], | ||
| "autopilot": [], | ||
| "unknown": [], | ||
| } | ||
|
|
@@ -784,6 +847,7 @@ def _summary_metrics_contract(buckets: dict[str, list[dict[str, Any]]]) -> dict[ | |
| buckets["verifier_followup_ledger"], | ||
| ), | ||
| "autopilot": _summarise_autopilot(buckets["autopilot"]), | ||
| "codex_cli_freshness": _summarise_codex_cli_freshness(buckets["codex_cli_freshness"]), | ||
| "unknown": {"records": len(buckets["unknown"])}, | ||
| } | ||
| ) | ||
|
|
@@ -947,6 +1011,7 @@ def build_summary( | |
| buckets["verifier_followup_ledger"], | ||
| ) | ||
| autopilot = _summarise_autopilot(buckets["autopilot"]) | ||
| codex_cli_freshness = _summarise_codex_cli_freshness(buckets["codex_cli_freshness"]) | ||
|
|
||
| now = _dt.datetime.now(_dt.UTC).replace(microsecond=0).isoformat().replace("+00:00", "Z") | ||
| lines = [ | ||
|
|
@@ -959,6 +1024,7 @@ def build_summary( | |
| f"verifier {verifier['runs']}, " | ||
| f"terminal dispositions {verifier['terminal_records']}, " | ||
| f"verifier follow-up ledgers {verifier['ledger_records']}, " | ||
| f"codex CLI freshness {codex_cli_freshness['records']}, " | ||
| f"autopilot {autopilot['records']}, " | ||
| f"unknown {len(buckets['unknown'])})" | ||
| ), | ||
|
|
@@ -1019,6 +1085,7 @@ def build_summary( | |
| f"{verifier['ledger_policy_depth_limit_exceeded']}" | ||
| ), | ||
| f"- Verifier models: {_format_counter(verifier['verifier_models'])}", | ||
| f"- Verifier CLI versions: {_format_counter(verifier['verifier_cli_versions'])}", | ||
| f"- Unsupported verifier models: {_format_counter(verifier['unsupported_verifier_models'])}", | ||
| ( | ||
| "- Unsupported model dispositions: " | ||
|
|
@@ -1034,6 +1101,25 @@ def build_summary( | |
| ), | ||
| f"- Model selection reasons: {_format_counter(verifier['model_selection_reasons'])}", | ||
| f"- Verifier modes: {_format_counter(verifier['verifier_modes'])}", | ||
| "", | ||
| "## Codex CLI Freshness", | ||
| f"- Records: {codex_cli_freshness['records']}", | ||
| f"- Statuses: {_format_counter(codex_cli_freshness['statuses'])}", | ||
| f"- Packages: {_format_counter(codex_cli_freshness['packages'])}", | ||
| f"- Pinned versions: {_format_counter(codex_cli_freshness['pinned_versions'])}", | ||
| f"- Latest versions: {_format_counter(codex_cli_freshness['latest_versions'])}", | ||
| f"- Outdated records: {codex_cli_freshness['outdated_records']}", | ||
| ( | ||
| "- Latest unavailable records: " | ||
| f"{codex_cli_freshness['latest_unavailable_records']}" | ||
| ), | ||
| ( | ||
| "- Max version delta: " | ||
| f"major {codex_cli_freshness['max_version_delta']['major']}, " | ||
| f"minor {codex_cli_freshness['max_version_delta']['minor']}, " | ||
| f"patch {codex_cli_freshness['max_version_delta']['patch']}" | ||
| ), | ||
| f"- Update targets: {_format_counter(codex_cli_freshness['update_targets'])}", | ||
| ] | ||
| ) | ||
|
|
||
|
|
||
Oops, something went wrong.
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.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
compactSelectionDetails()storesselection.statusviacleanString(selection.status || 'pass')without normalizing casing. Since later logic treats'error'as a sentinel (buildInitialManifest/finalizeManifestcompare to'error'), a non-lowercase value like'Error'would silently fail to propagate an error status. Consider normalizing this field (e.g.,cleanString(...).toLowerCase()or reusenormalizeStatus()with allowed values).