feat(dashboard): Active Observations surface — page 27 + System Health KPI strip#135
Merged
Merged
Conversation
…h KPI strip
Phase 2 of the observation-registry arc. Companion to the
/Artifact_Freshness surface (page 26) — that page covers the freshness
axis ("does the artifact land on time?"); this one covers the
observation axis ("is the consumer plumbed to read it yet, and what's
the cutover gate?").
Per feedback_observe_mode_unconditional_gates_govern_cutover: observe-
mode producer code runs unconditionally; activation flags gate ONLY
the consumer-cutover transition. This surface makes the gates visible
so a soak that passes its acceptance criterion isn't silently forgotten.
Changes:
- `loaders/observation_registry_loader.py` — direct-from-disk YAML
loader. No S3 hop / no Lambda dependency (the registry IS the data).
Path resolution: env override → EC2 console path
(/home/ec2-user/alpha-engine-config/...) → local-dev path
(~/Development/alpha-engine-config/...) → sibling-directory fallback.
Applies the YAML's `defaults` block to each entry (so
`verification_status: audit-found-needs-curation` survives without
per-row repetition). Returns None if no path resolves; caller
renders a "registry not found" panel rather than crashing.
- `pages/27_Active_Observations.py` — deep-dive surface mirroring
page 26's structure: KPI strip (counts by state + phase), filters
(state / phase / producer_repo / sort), table view, per-entry
expanders showing the full cutover_gate / evidence / composes_with
detail.
- `pages/4_System_Health.py` — adds Section 0.5 "Active Observations"
KPI strip directly below the existing Section 0 freshness strip.
Same 7-column layout: total / always-on / gated-on / gated-off /
substrate / cutover / promoted. Links to /Active_Observations for
the deep dive.
Live smoke (local):
- Loader resolves to ~/Development/alpha-engine-config/private-docs/...
- 13 observations loaded, defaults merge confirmed
- All 13 entries carry the keys the page reads
- Streamlit server starts clean, HTTP 200 on the page route, no
tracebacks in the server log
Deploy: per CLAUDE.md, code-only PRs need explicit restart on both
console + public services. After merge:
ae-dashboard "sudo systemctl start boot-pull && \
sudo systemctl restart dashboard && \
sudo systemctl restart nous-ergon-public"
boot-pull.sh will refresh both the dashboard repo AND the
alpha-engine-config clone the loader reads.
Composes with:
- alpha-engine-config #349 (seed registry, 2026-05-28)
- alpha-engine-config #351 (validator + GHA chokepoint)
- alpha-engine-data #338 (CheckEnableStandaloneScanner removal)
- feedback_observe_mode_unconditional_gates_govern_cutover (2026-05-28)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Deploying nousergon-marketing with
|
| Latest commit: |
e763cdd
|
| Status: | ✅ Deploy successful! |
| Preview URL: | https://2616b5b9.nousergon-marketing.pages.dev |
| Branch Preview URL: | https://feat-active-observations-sur.nousergon-marketing.pages.dev |
cipher813
added a commit
that referenced
this pull request
May 28, 2026
…on artifacts
Closes the cross-registry-visibility gap surfaced by the operator
question "are the observation-only artifacts marked as such in the
artifact freshness page?" Severity (warning/critical) was overloaded
and there was no way to tell which warning entries were observe-mode
rollouts vs which were secondary observability for production runs.
Derivation rule (single SoT, no new data — uses cross-link in
OBSERVATION_REGISTRY.yaml + severity in ARTIFACT_REGISTRY.yaml):
- production: severity == critical OR no active observation entry's
composes_with references this artifact
- observation: severity == warning AND at least one active
(state in {always-on, gated-on}) observation references this id
Live verification: 51 entries → 3 observation + 48 production.
- observation: scanner_candidates_json, factor_profiles_latest,
research_scorecard_latest
- production: includes critical entries (signals.json,
predictor_predictions, weekly_collector_manifest) AND warning-
severity secondary-observability (macro_snapshot,
research_consolidated_morning)
- Boundary case: signals.json is severity=critical AND observed by
attractiveness_pillars_phase_4 → classified production (critical
wins; cross-link doesn't override load-bearing status)
Changes:
- Imports observation_registry_loader (shipped in #135)
- New _load_active_observation_artifact_refs helper (TTL 60s cache)
- _derive_type applies the rule per-row
- df.apply builds a type column
- New "Type" filter (5th multiselect)
- New "Type" column inserted between State and Artifact in display table
Deploy after merge:
ae-dashboard "sudo systemctl start boot-pull && sudo systemctl restart dashboard"
Composes with alpha-engine-config #349/#351/#352/#355 (registry arc)
+ alpha-engine-dashboard #135 (page 27) + #136 (format_age fix) +
[[feedback_observe_mode_unconditional_gates_govern_cutover]].
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
cipher813
added a commit
that referenced
this pull request
May 28, 2026
…on artifacts (#137) Closes the cross-registry-visibility gap surfaced by the operator question "are the observation-only artifacts marked as such in the artifact freshness page?" Severity (warning/critical) was overloaded and there was no way to tell which warning entries were observe-mode rollouts vs which were secondary observability for production runs. Derivation rule (single SoT, no new data — uses cross-link in OBSERVATION_REGISTRY.yaml + severity in ARTIFACT_REGISTRY.yaml): - production: severity == critical OR no active observation entry's composes_with references this artifact - observation: severity == warning AND at least one active (state in {always-on, gated-on}) observation references this id Live verification: 51 entries → 3 observation + 48 production. - observation: scanner_candidates_json, factor_profiles_latest, research_scorecard_latest - production: includes critical entries (signals.json, predictor_predictions, weekly_collector_manifest) AND warning- severity secondary-observability (macro_snapshot, research_consolidated_morning) - Boundary case: signals.json is severity=critical AND observed by attractiveness_pillars_phase_4 → classified production (critical wins; cross-link doesn't override load-bearing status) Changes: - Imports observation_registry_loader (shipped in #135) - New _load_active_observation_artifact_refs helper (TTL 60s cache) - _derive_type applies the rule per-row - df.apply builds a type column - New "Type" filter (5th multiselect) - New "Type" column inserted between State and Artifact in display table Deploy after merge: ae-dashboard "sudo systemctl start boot-pull && sudo systemctl restart dashboard" Composes with alpha-engine-config #349/#351/#352/#355 (registry arc) + alpha-engine-dashboard #135 (page 27) + #136 (format_age fix) + [[feedback_observe_mode_unconditional_gates_govern_cutover]]. Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
3 tasks
cipher813
added a commit
that referenced
this pull request
May 28, 2026
Surfaces the per-cycle history for each artifact, reading the new
_freshness_monitor/history.json written daily at 04:00 UTC by the
freshness-monitor Lambda's historical mode (alpha-engine-data PR #339).
Closes the gap surfaced 2026-05-28: 'are there gaps in the producer's
history?' — operators want to know not just current-cycle state but
whether last weekend / last month had silent absences.
Changes:
- _load_history loader (TTL 300s — refreshes once/day, not 15min)
- New History (12wk) column on the main table:
✅ N/N continuous — clean history
⚠️ G/N gaps — gappy producer
✅ exists (latest) — latest-pointer present
❌ absent (latest) — latest-pointer missing
— — historical probe hasn't covered this id yet
(continuous-cadence artifacts skip historical mode)
- New 'Per-artifact history drill-down' section below the main
table. Each artifact in the filtered view gets an expander
showing the per-cycle sequence (date / present / size /
last_modified / error_code). Sort: gappy first, continuous last;
latest-pointer absent at top, latest-pointer present at bottom.
First 3 worst-offender entries auto-expand.
- Graceful-degrade: if history.json doesn't exist yet, page shows
a single info box explaining the daily cron + manual-invoke
instructions.
Operator caveat: calendar-naive. NYSE holidays may render as
false-positive ❌ absent cells. Calendar-aware probe is a future
enhancement (P3 in the Lambda PR).
Composes with alpha-engine-data PR #339 (historical-mode Lambda)
+ the prior page 26/27 work in #134/#135/#136/#137.
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.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
Phase 2 of the observation-registry arc (Phase 1: alpha-engine-config #349 seed + #351 validator). Companion surface to /Artifact_Freshness (page 26).
loaders/observation_registry_loader.pyreadsOBSERVATION_REGISTRY.yamldirectly from local disk.pages/27_Active_Observations.py— deep-dive surface with filters, table, and per-entry expanders.pages/4_System_Health.py— adds Section 0.5 KPI strip below the existing Section 0 freshness strip.Why
Per
feedback_observe_mode_unconditional_gates_govern_cutover(2026-05-28): observe-mode producer code runs unconditionally; activation flags gate ONLY the consumer-cutover transition. This surface makes the gates visible so a soak that passes its acceptance criterion isn't silently forgotten.Sibling axis to page 26: freshness asks "does the artifact land on time?"; observation asks "is the consumer plumbed to read it yet?"
Architecture
The registry IS the data — no Lambda or S3 hop. The dashboard EC2 instance already pulls
alpha-engine-configviainfrastructure/boot-pull.sh(sibling repo at/home/ec2-user/alpha-engine-config). The loader walks four candidate paths: env override → EC2 console path → local-dev path → sibling-directory fallback.If the file isn't resolvable, both the page and the System Health KPI strip degrade gracefully (page shows an error panel, KPI strip is skipped — same pattern the existing freshness KPI strip uses when the heartbeat S3 key is absent).
Page 27 layout
Test plan
summarize_by_state/summarize_by_phasereturn correct counts (2 always-on / 4 gated-on / 7 gated-off; 1 substrate / 10 observe / 1 cutover / 1 promoted)Deploy after merge
Per CLAUDE.md, code-only PRs need explicit restart on both the console + public services:
```
ae-dashboard "sudo systemctl start boot-pull && \
sudo systemctl restart dashboard && \
sudo systemctl restart nous-ergon-public"
```
boot-pull.shrefreshes both the dashboard repo AND thealpha-engine-configclone the loader reads.Composes with
feedback_observe_mode_unconditional_gates_govern_cutover(2026-05-28)🤖 Generated with Claude Code