Add enterprise dashboard accessibility guard

[KoiosSG]

Portfolio Comparison Refresh (2026-06-27)

• Current live state: open, non-draft, clean mergeability on head 86978f2; no GitHub check runs or status contexts are attached, and Algora remains Pending with Total paid $0.
• Same-issue distinction: the latest known same-issue competitor PR Add enterprise token rotation guard #586 targets enterprise token rotation, while this PR targets enterprise dashboard accessibility and private-data-safe admin evidence before dashboard release, scheduled export, or webhook summary.
• Reviewer evidence advantage: this PR covers contrast evidence, missing/invalid/unresolved CSS colors, shorthand hex handling, screen-reader labels, keyboard reachability, focus traps, visible focus indicators, private-data scans in accessibility text, table summaries, heading order, reduced motion, malformed component evidence, JSON/Markdown/SVG reports, and MP4 reviewer evidence.

/claim #19 ## Summary - Adds a self-contained enterprise-dashboard-accessibility-guard/ slice for Enterprise Tooling issue #19. - Evaluates institutional admin dashboard releases before they are shown to admins, included in scheduled exports, or summarized through webhook notices. - Checks contrast for critical and noncritical metrics, missing/invalid/unresolved contrast evidence, shorthand hex colors, screen-reader labels, keyboard reachability, focus traps, visible focus indicators, private-data exposure in accessibility text, table summaries, heading order, reduced-motion fallbacks, malformed component evidence, and malformed top-level dashboard packets. - Emits deterministic JSON, Markdown, SVG, and MP4 reviewer artifacts with SHA-256 audit digests. ## Hardening Updates - Noncritical low-contrast dashboard content now produces a warning, keeps scheduled exports blocked, and marks WCAG perceivable readiness false instead of releasing as clean. - Table/export accessibility summaries are included in the private-data scan, with explicit redact_accessibility_text:* remediation. - Invalid or unresolved contrast evidence, including CSS color tokens such as var(--metric-danger), now blocks dashboard release instead of being parsed into a false clean color. - Standard shorthand hex contrast evidence such as #000 / #fff remains valid, so the guard is strict about invalid evidence without over-blocking common CSS syntax. - Keyboard-reachable dashboard controls that suppress visible focus indicators now block release instead of being marked accessibility-ready. - Critical dashboard widgets that omit foreground/background contrast evidence entirely now block release with INVALID_CONTRAST_EVIDENCE and provide_valid_contrast_evidence:* remediation. - Noncritical dashboard metrics that omit contrast evidence now enter remediation before public release, keep scheduled exports blocked, and mark WCAG perceivable readiness false. - Malformed dashboard component evidence, such as widgets: [null], now blocks release with MALFORMED_DASHBOARD_COMPONENT_ENTRY and repair_dashboard_component_evidence:* instead of crashing assessment before reviewer packets are generated. - Malformed top-level dashboard packets, such as assessDashboardRelease(null), now block release with MALFORMED_DASHBOARD_PACKET, stable unidentified-dashboard packet identity, and repair_dashboard_packet:* remediation instead of crashing before reviewer packets are generated. - Malformed reduced-motion evidence, such as motion.animatedCharts supplied as a string, now blocks release with MALFORMED_MOTION_EVIDENCE and repair_motion_evidence:* instead of crashing animated chart assessment. ## Scope This focuses specifically on accessibility readiness for institutional admin dashboards and their downstream export/webhook release lanes. It is distinct from the existing dashboard/export/webhook replay/compliance/identity/retention/data-residency/SLA/secret-rotation/quota/API-change/connector-certification/incident/funder/AI-model/dashboard-attribution/initiative-tag/policy-exception/IRB/data-export/SCIM/deposit-reconciliation/admin-notification/cost-allocation/LMS/payload-redaction/vendor-DPA/cohort-privacy/API-rate-limit/training-policy slices. ## Validation - Red regression on the parent of a0f3773 reproduced the malformed motion crash first: TypeError: dashboard.motion.animatedCharts.map is not a function. - Red regression on the parent of 6f06509 reproduced the malformed top-level dashboard crash first: TypeError: Cannot read properties of null (reading 'widgets'). - Red regression on the parent of c2eae84 reproduced the malformed component crash first: TypeError: Cannot read properties of null (reading 'critical'). - cd enterprise-dashboard-accessibility-guard && npm test -> 13 tests passed. - cd enterprise-dashboard-accessibility-guard && npm run demo -> regenerated blocked, missing-contrast, missing-noncritical-contrast, malformed-component, malformed-dashboard, malformed-motion, clean, and warning JSON/Markdown/SVG artifacts. - cd enterprise-dashboard-accessibility-guard && npm run demo:video -> regenerated reports/demo.mp4. - cd enterprise-dashboard-accessibility-guard && npm run check -> passed JS syntax checks and Python compile check. - Generated JSON packets parsed successfully: 8 packets. - ffprobe confirmed reports/demo.mp4 is H.264, 1280x720, 24fps, 9.041667s, 132,119 bytes. - git diff --check and git diff --cached --check passed; Git only reported Windows line-ending normalization warnings. - Focused restricted-string scan returned 0 matches. - GitHub PR merge state after push confirmation: CLEAN; no checks are reported for this branch. ## Safety Synthetic data only. No credentials, private dashboard records, SSO calls, webhook calls, export delivery, payment processor calls, or private institutional systems are used. AI-assisted with OpenAI Codex; I reviewed and locally verified the diff before submitting.

[KoiosSG]

Hardening update pushed in 5047577: noncritical low-contrast dashboard content now produces a warning, keeps scheduled exports blocked, and marks WCAG perceivable readiness false instead of releasing as clean. I added a regression that failed before the fix with release_with_accessibility_monitoring == remediate_before_public_release and now passes. Validation refreshed locally: npm run check, npm test (4 tests), npm run demo, npm run demo:video, ffprobe on demo.mp4, git diff --check, and credential-like sensitive-term scan returned no matches.

[KoiosSG]

Hardening update pushed in 836537b: table/export accessibility summaries are now included in the private-data scan, and private accessibility text produces an explicit redact_accessibility_text:* remediation action. This closes a leak path where a screen-reader label could be safe while a table summary still exposed a restricted project or private lab owner.

Verification refreshed:

• Red regression first: npm test failed on the new table-summary private-data case (release_with_accessibility_monitoring vs hold_accessibility_release).
• Green: npm test passes with 5 enterprise dashboard accessibility guard tests.
• npm run check passes syntax checks for JS and Python files.
• npm run demo regenerated JSON/Markdown/SVG artifacts; blocked packet now includes the redact accessibility text action.
• npm run demo:video regenerated reports/demo.mp4.
• ffprobe confirms reports/demo.mp4 is H.264, 1280x720, 24fps, 7.5s, 104,275 bytes.
• git diff --check and git diff --cached --check pass.
• Credential/payout-focused scan across changed code/docs/reports returned no matches.

[KoiosSG]

Follow-up competitive hardening pass for the enterprise dashboard accessibility guard.

What changed in e0195d7:

• Added a regression for invalid/unresolved contrast evidence such as CSS tokens (var(--metric-danger)) being treated as clean color evidence.
• Invalid contrast evidence now blocks dashboard release with INVALID_CONTRAST_EVIDENCE and provide_valid_contrast_evidence:* remediation.
• Added a companion regression so valid shorthand hex contrast evidence (#000 / #fff) remains accepted rather than over-blocked.
• Refreshed README, requirements map, acceptance notes, and reviewer artifacts; the blocked packet now includes the invalid-contrast-evidence finding.

Validation refreshed locally:

• Confirmed the CSS-token regression failed before implementation with release_with_accessibility_monitoring instead of hold_accessibility_release.
• Confirmed the shorthand-hex regression failed before support was added with hold_accessibility_release instead of release_with_accessibility_monitoring.
• npm test -> 7 enterprise dashboard accessibility guard tests passed.
• npm run demo -> regenerated JSON/Markdown/SVG artifacts; blocked packet findings now include INVALID_CONTRAST_EVIDENCE.
• npm run demo:video -> regenerated reports/demo.mp4.
• npm run check -> JS syntax checks and Python compile check passed.
• ffprobe verified reports/demo.mp4 as H.264, 1280x720, 24 fps, 7.5s, 104,275 bytes.
• git diff --check and git diff --cached --check passed; only Git line-ending normalization warnings appeared on Windows.
• Credential/payout-focused scan returned no matches.

[KoiosSG]

Follow-up competitive hardening pass for the enterprise dashboard accessibility guard.

What changed in 69c962d:

• Added a regression for keyboard-reachable dashboard controls that suppress visible focus indicators.
• Such controls now block dashboard/export/webhook release with MISSING_VISIBLE_FOCUS_INDICATOR instead of being marked accessibility-ready.
• The remediation packet now emits add_visible_focus_indicator:* so reviewers can see the exact required accessibility action.
• README, requirements map, acceptance notes, and generated reviewer artifacts were refreshed; the blocked packet now includes the visible-focus finding.

Validation refreshed locally:

• Confirmed the new regression failed before implementation with release_with_accessibility_monitoring instead of hold_accessibility_release.
• npm test -> 8 enterprise dashboard accessibility guard tests passed.
• npm run check -> JS syntax checks and Python compile check passed.
• npm run demo -> regenerated JSON/Markdown/SVG artifacts; blocked packet now has 10 findings including MISSING_VISIBLE_FOCUS_INDICATOR.
• npm run demo:video -> regenerated reports/demo.mp4.
• node --check on index/sample-data/test/demo passed.
• ffprobe verified reports/demo.mp4 as H.264, 1280x720, 24 fps, 7.5s, 104,275 bytes.
• git diff --check and git diff --cached --check passed.
• Credential/payout-focused scan returned no actual secrets or payout strings; expanded-pattern hits were source/doc words such as color token and the README safety sentence.
• GitHub PR merge state after push: CLEAN.

[KoiosSG]

Follow-up competitive hardening pass for the enterprise dashboard accessibility guard.

What changed in 13606f9:

• Added a regression for critical dashboard metrics that omit foreground/background contrast evidence entirely.
• Critical widgets now block release with INVALID_CONTRAST_EVIDENCE unless both foreground and background color evidence are present and parseable.
• Added reports/missing-contrast-packet.json and refreshed README, requirements, acceptance notes, Markdown/SVG reports, and the demo video so reviewers can inspect the new path.

Why this matters:

• A critical institutional dashboard metric without contrast evidence is not accessibility-ready just because no invalid color token was supplied.
• This closes a release-gating bypass where the dashboard could be marked release_with_accessibility_monitoring without evidence that critical visual content is perceivable.

Validation refreshed locally:

• Confirmed the new regression failed before implementation with release_with_accessibility_monitoring instead of hold_accessibility_release.
• npm test -> enterprise dashboard accessibility guard tests passed (9).
• npm run demo -> regenerated blocked/missing-contrast/clean/warning packets; missing-contrast-packet.json blocks with INVALID_CONTRAST_EVIDENCE.
• npm run demo:video -> regenerated reports/demo.mp4.
• npm run check -> JS syntax checks and Python compile check passed.
• ffprobe verified reports/demo.mp4 as H.264, 1280x720, 24 fps, 7.5s, 104,275 bytes.
• git diff --check and git diff --cached --check passed; Git only reported Windows line-ending normalization warnings.
• Focused sensitive scan found no actual secrets or payout strings; only safety/scope wording matched the broad pattern.
• GitHub PR merge state after push: CLEAN; no checks are reported for this branch.

[KoiosSG]

Follow-up competitive hardening pass for the enterprise dashboard accessibility guard.

What changed in 0a193d2:

• Added a regression for noncritical dashboard metrics that omit foreground/background contrast evidence entirely.
• Noncritical missing contrast evidence now produces warning-level INVALID_CONTRAST_EVIDENCE, keeps public release in remediation, blocks scheduled exports, and marks WCAG perceivable readiness false instead of releasing clean.
• Added reports/missing-noncritical-contrast-packet.json and regenerated Markdown/SVG/MP4 reviewer evidence.

Why this matters:

• Noncritical dashboard content still needs valid contrast evidence before public/admin release. Otherwise a metric can bypass the existing noncritical low-contrast warning path simply by omitting the evidence.

Validation refreshed locally:

• Confirmed the new regression failed before implementation with release_with_accessibility_monitoring instead of remediate_before_public_release.
• npm test -> enterprise-dashboard-accessibility-guard tests passed (10).
• npm run demo -> regenerated 5 JSON packets plus Markdown/SVG evidence.
• npm run demo:video -> regenerated reports/demo.mp4.
• npm run check -> JS syntax checks and Python compile check passed.
• ffprobe verified reports/demo.mp4 as H.264, 1280x720, 7.5s, 180 frames.
• All generated JSON packets parsed successfully.
• git diff --check and git diff --cached --check passed.
• Focused restricted-string scan returned no matches.

[KoiosSG]

Follow-up competitive hardening pass for the enterprise dashboard accessibility guard.

What changed in c2eae84:

• Added a regression for malformed dashboard component entries such as widgets: [null] crashing release assessment before packet generation.
• Malformed component evidence now blocks release with MALFORMED_DASHBOARD_COMPONENT_ENTRY and emits repair_dashboard_component_evidence:* remediation.
• Added reports/malformed-component-packet.json and refreshed Markdown/SVG/MP4 reviewer evidence.

Validation refreshed locally:

• Confirmed the new regression failed before implementation with TypeError: Cannot read properties of null (reading 'critical').
• npm test -> enterprise-dashboard-accessibility-guard tests passed (11).
• npm run check, npm run demo, and npm run demo:video passed.
• Generated JSON packet parse check passed for 6 packets.
• ffprobe verified reports/demo.mp4 as H.264, 1280x720, 24fps, 9.04s, 132,119 bytes.
• git diff --check and git diff --cached --check passed.
• Focused restricted-string scan found no high-confidence secrets, tokens, or payout strings.
• GitHub PR merge state after push: CLEAN; no checks are reported for this branch.

[KoiosSG]

Follow-up hardening verified in c2eae84:

• Malformed dashboard component evidence such as widgets: [null] now blocks release with MALFORMED_DASHBOARD_COMPONENT_ENTRY and repair_dashboard_component_evidence:* instead of crashing before reviewer packets are generated.
• Red check against the parent commit reproduced TypeError: Cannot read properties of null (reading 'critical') before the fix.
• Verification refreshed: npm test (11), npm run demo, npm run demo:video, npm run check, 6 generated JSON packets parsed, ffprobe confirmed H.264 1280x720 24fps 9.041667s / 132,119 bytes, diff checks passed, and high-confidence secret/payout scan returned 0 matches.
• GitHub PR merge state after push confirmation: CLEAN; no checks are reported for this branch.

[KoiosSG]

Follow-up hardening pushed in 6f06509 for malformed top-level dashboard packets.

What changed:

• assessDashboardRelease(null) now returns a deterministic blocker packet instead of crashing at dashboard.widgets.
• The new packet uses stable unidentified-dashboard identity, emits MALFORMED_DASHBOARD_PACKET, and routes repair_dashboard_packet:unidentified-dashboard remediation before dashboard/export/webhook release can proceed.
• Added reports/malformed-dashboard-packet.json plus refreshed Markdown/SVG reviewer evidence.

Verification refreshed locally:

• Red regression first reproduced TypeError: Cannot read properties of null (reading 'widgets').
• npm test -> 12 tests passed.
• npm run check, npm run demo, and npm run demo:video passed.
• 7 generated JSON packets parsed successfully.
• ffprobe confirmed H.264 1280x720 24fps 9.041667s / 132,119 bytes.
• git diff --check, git diff --cached --check, staged allowlist check, and focused restricted-string scan passed.
• GitHub PR merge state after push: CLEAN; no checks are reported for this branch.

[KoiosSG]

Follow-up hardening pushed in a0f3773 for malformed reduced-motion evidence.

What changed:

• motion.animatedCharts must now be an array of animated component IDs before reduced-motion assessment runs.
• Malformed motion evidence such as a string now blocks release with MALFORMED_MOTION_EVIDENCE and repair_motion_evidence:motion.animatedCharts instead of crashing at .map() before reviewer packets are generated.
• Added reports/malformed-motion-packet.json and refreshed Markdown/SVG/MP4 reviewer evidence.

Verification refreshed locally:

• Red regression first reproduced TypeError: dashboard.motion.animatedCharts.map is not a function.
• npm test -> 13 tests passed.
• npm run check, npm run demo, and npm run demo:video passed.
• 8 generated JSON packets parsed successfully.
• ffprobe confirmed reports/demo.mp4 as H.264, 1280x720, 24fps, 9.041667s / 132,119 bytes.
• git diff --check, git diff --cached --check, staged allowlist check, and focused restricted-string scan passed.
• GitHub PR merge state after push: CLEAN; no checks are reported for this branch.

[KoiosSG]

Hi, just checking in on this bounty submission. The enterprise dashboard accessibility guard is still open and merge-clean; the current package covers malformed dashboard packets/components, reduced-motion evidence, contrast evidence, and identifier redaction with regenerated reviewer artifacts. Is there anything specific you would like me to change, simplify, or clarify to make review/selection easier?