| 2026-08-28 | Grok Bot harness: pass the Cursor Marketplace review gate (self-contained logo, displayName, license) | plugins mcp discovery |
| 2026-08-28 | Integrator DX: SDK recovery, unified error envelope, /v1/mcp guidance, load_shed 429 | api sdk mcp routing |
| 2026-08-28 | Convergence wave: bind provenance on the real audit path; dashboard favicon + honest sub-cent cost | governance dashboard cdn |
| 2026-08-28 | Stochastic review round 2: 21 defects fixed, alerting resurrected, false attestations killed | |
| 2026-08-28 | Stochastic review round 3: honest savings, closed explain loops, first-class model substitution | routing governance cost-attribution observability mcp tenant-onboarding dashboard |
| 2026-08-28 | Stochastic review round 4: the data-residency false guarantee, closed for real | data-protection routing mcp cost-attribution governance rbac dashboard |
| 2026-08-28 | Round-6 remediation: force provenance binding, honest cache accounting, x-br-build everywhere | governance cost-attribution data-protection observability |
| 2026-08-28 | Stochastic review round 4: fully-audited re-score (7.5→8.2), then 13 defects closed | |
| 2026-08-28 | Stochastic review round 5: a fix wave regressed observability — caught, root-caused, and closed | |
| 2026-08-27 | BYOK graduation: an agent goes from nothing to standard tier with zero humans | tenant-onboarding community-tier mcp-server billing |
| 2026-08-27 | Learning loop closed at all three joints — truthful labels, streaming trace, outcome taxonomy | router intelligence governance observability |
| 2026-08-27 | Marketing overhaul: one positioning — the governed control plane for AI agents | router-intelligence governance capability-graph mni a2a-mesh |
| 2026-08-27 | curl brainstormrouter.com now answers the agent — and br go onboards in one shot | edge-worker cli tenant-onboarding |
| 2026-08-26 | Grok Bot harness: BrainstormRouter as the Bot's control plane | mcp-server capability-graph routing memory |
| 2026-08-26 | Comparative router harness — capability + performance vs Portkey / OpenRouter / Cursor / DigitalOcean | router intelligence |
| 2026-08-26 | Production hardened, stress-tested, and audit-reviewed — infra idle under 90 req/s flood, $0/mo added | router intelligence governance observability |
| 2026-08-12 | BR self-describes: a machine-queryable, trust-scored Capability Graph (query, don't grep) | api capabilities mni capability-graph |
| 2026-08-12 | Self-service processor-fact attestation — a constrained request can finally route | data-protection security router |
| 2026-08-11 | auto routing rebuilt: capability-matched, learning, and on for every tier | router model-auto-selector community-tier |
| 2026-08-11 | Self-reporting honesty pass: advertised catalog matches the live endpoint, header matches body, timestamps agree | router api security discovery |
| 2026-08-11 | Comparative router harness — capability + performance vs Portkey / OpenRouter / Cursor / DigitalOcean | router intelligence |
| 2026-08-10 | Comparative router harness — capability + performance vs Portkey / OpenRouter / Cursor / DigitalOcean | router intelligence |
| 2026-08-10 | Router uplevel — bank the leads, make everything observable, address the one gap | router api intelligence security |
| 2026-08-09 | Round-2 remediation Wave 1 — BYOK honesty, cache-governance, tenant-scoped intelligence, honest self-description, live MCP agent tools | router governance security api mcp sdk |
| 2026-08-09 | Round-3 remediation — durable RMM memory (Postgres) + cache-hit request-id so cache-served responses are auditable | memory governance router api |
| 2026-08-09 | Round-4 remediation — model-contract honesty + memory injection on the completions path | router memory api |
| 2026-08-09 | Round-4 remediation — cost truthfulness + routing explain/health honesty | router intelligence api sdk |
| 2026-08-09 | Fix first-copy-paste 403 dead-end + derive catalog counts from source (was hand-copied, 3-way drift) | sdk-ts sdk-py docs ai-discoverability |
| 2026-08-09 | Stochastic Review convergence wave — count truthfulness, cache-hit lineage, savings honesty, data-policy provability, alert deliverability | router api intelligence security infra sdk |
| 2026-08-09 | Round-6 remediation: 7 blockers closed — priced explain alternatives, request-level data-protection enforcement, evidence-gated cache, honest savings | router api security observability intelligence |
| 2026-08-08 | Closed the cache bypass of provider.only — caches now carry processor provenance | router semantic-cache prompt-cache data-protection |
| 2026-08-08 | Evidence you can actually verify: RFC 8785 canonicalization + Ed25519 signatures + per-check verify | governance security audit-chain |
| 2026-08-08 | Two guardrail lies fixed: a config write that reported success and did nothing, a summary that asserted a scan it never ran | security guardrails governance |
| 2026-08-08 | Closed a pooled-inference leak: anonymous tenants were billing real model calls to the platform | api agents security |
| 2026-08-08 | provider.only / provider.ignore are now enforced — the data-protection control that did nothing | router completions-api explain |
| 2026-08-08 | RBAC registry drift — governance.write orphan, audit/sign downgrade, masked 404s | rbac governance |
| 2026-08-08 | Six defects closed from one review round — including a headline feature that was 100% broken | router api mcp budget infra gateway |
| 2026-08-08 | One authority per data class: five 'reads zero' bugs and a hard 500 were all the same split-brain | security governance api billing router |
| 2026-08-08 | Six checks that reported success while doing nothing — 27 days of undeployed code, recovered | infra deploy security media |
| 2026-08-08 | Stochastic Review Round 2 — 10 personas, audited mean 5.4 on a pinned build | router governance security observability api |
| 2026-08-07 | Evidence-required mode, policy-attributed artifacts, and an honest cache key | router security governance api |
| 2026-08-05 | Increment 0A — removed five false compliance guarantees from the request path | router security governance |
| 2026-08-05 | Routing constraints are now inviolable — data-policy and provider-deny survive filter starvation | router governance security |
| 2026-08-04 | Telemetry truthfulness: three fields that reported things that weren't true | router api |
| 2026-08-01 | Model catalog was silently truncating responses to 25% of capacity — fixed, with a regression test | router provider-catalog |
| 2026-07-24 | Governance MCP tools — br_get_governance_artifact + br_list_policy_packs land the two agent-facing follow-ups | mcp security governance decision-lineage |
| 2026-07-21 | Evidence coverage metric — the demo number that reframes governance from 'less manual work' to '~100% bound' | security governance decision-lineage |
| 2026-07-21 | Governance Artifact — signed, control-mapped, verifiable evidence assembled from a real inline decision | security governance decision-lineage audit-chain |
| 2026-07-21 | Policy Pack Library — named, framework-tagged governance packs that enforce inline | security policy-engine governance |
| 2026-07-20 | One runtime governance decision plane + an open policy-as-code SDK — Phase 1 of out-innovating Credo AI | security policy-engine guardrails mcp tool-call-firewall |
| 2026-07-11 | Agent-native onboarding — zero to first routed completion in 2 API calls, no invite, no email, no human | onboarding register signup community-tier provisioning |
| 2026-07-04 | Add GLM-5.2 (Z.ai) as a first-class provider — frontier-class open weights at 1/6th the cost | router provider-catalog dashboard |
| 2026-06-28 | BrainstormRouter runs live on AWS Fargate for ~$0 incremental — full routing for ~$28/mo | router gateway infra |
| 2026-06-21 | Increment 5 — Reposition: 'The authorization & evidence plane for AI agents' | capability-grants scope-enforcement audit-chain decision-lineage mcp-server |
| 2026-06-21 | Track B Phase 2 gate — JWT-human grant read/simulate/revoke (the CR-grants block) | capability-grants scope-enforcement audit-chain governance-dashboard |
| 2026-06-21 | Track B Phase 2 frontend — the #grants surface (read · simulate · revoke) | governance-dashboard capability-grants |
| 2026-06-21 | Track B Phase 3 (CR-1) — chain-verify badge over a shared verify service | audit-chain governance-dashboard scope-enforcement |
| 2026-06-21 | Track B Phase 3 (CR-1) — the chained Ledger view (+ head hash) | audit-chain governance-dashboard |
| 2026-06-21 | Track B Phase 3 (CR-2) — configured-vs-effective scope enforcement (cohort divergence) | scope-enforcement governance-dashboard feature-flags |
| 2026-06-21 | Track B Phase 3 (CR-3) — would-deny preview (warn-mode, deduped, lookback) | scope-enforcement audit-chain governance-dashboard |
| 2026-06-21 | Track B Phase 3 (CR-4) — dry multi-source scope composition | scope-enforcement governance-dashboard |
| 2026-06-21 | Track B Phase 3 (CR-5) — effective-state badge (kill-switch override, honest) | scope-enforcement governance-dashboard |
| 2026-06-21 | Track B Phase 3 (CR-7) — governance event types on the live feed | governance-dashboard capability-grants scope-enforcement sse |
| 2026-06-21 | Track B Phase 3 — live governance feed (dashboard consumes the CR-7 events) | governance-dashboard sse |
| 2026-06-21 | Track B Phase 4 (CR-9) — the Plans section (request-scoped, distinct from grants) | capability-grants liaison governance-dashboard |
| 2026-05-24 | Behavioral Contract Layer + arrival manifest: BR becomes legible to a cold frontier model | api capability-registry discovery |
| 2026-05-22 | Evidence Toolkit: Build Identity + Load-Shed Gate Closers | api sdk mcp load-shedding operations |
| 2026-05-22 | Load shedding: 100-VU 504 cascade → cheap 503 + Retry-After at the edge | api-middleware api-capabilities |
| 2026-05-22 | Single-session 19-PR + 2 R40 follow-up = 21-PR marathon: R37→R39 +1.13 score, all R39 items + 3/4 R40 items closed | |
| 2026-05-22 | R38/R39 session: 7 PRs closing load envelope, security audit, generator drift, single-region ADR, stream-drain primitive | api-server api-middleware build-pipeline ops-infra docs-adr |
| 2026-05-21 | R36 sprint: 4 PRs close 8 fortress fails, lift /v1/ops/status capacity signals | api-router api-middleware api-capabilities |
| 2026-05-20 | Quality-aware semantic cache admission — reject broken responses before they get cached | router cache intelligence |
| 2026-05-20 | Promote null-content detection to upstream 502 — cascade now retries instead of returning silent 200 | router completions intelligence |
| 2026-05-16 | /.well-known/build.json — single GET answers 'is this deploy current?' | api marketing-site ai-discoverability |
| 2026-05-16 | Open specs published: Trust Envelope v1 + MCP-OAuth Bridge v1 at docs.brainstormrouter.com/specs/ | docs specs sdks |
| 2026-05-14 | Trust envelope + ErrorEnvelope + activation gate — all 6 goal conditions live | trust-envelope error-envelope ops-status fortress |
| 2026-05-14 | Intelligence endpoints expose top-level .summary envelope (R25 probe-null fix) | api router intelligence |
| 2026-05-13 | Every route in src/api/routes/** now flows through defineCapability() — 618 routes, 0 inline registrations, hot path included | api capabilities completions auth |
| 2026-05-09 | A-2 — trust envelope synthesizer formalized: mtls_fingerprint, call counters, boot wiring | security trust-envelope gateway |
| 2026-05-09 | A-3 — trust envelope load-bearing for routing: tier/scope/xdr_risk/anomaly gates | router trust-envelope gateway |
| 2026-05-09 | Audit chain canonicalization — single source of truth fixes verify-side drift (was broken since 0d46a1728) | security api audit |
| 2026-05-09 | Dual-mount /v1/* for routing intelligence + ops endpoints — closes long-standing SDK drift | api sdk rbac |
| 2026-05-09 | Notification cooldown — move from emit-time bus gate to per-subscriber decoration | notifications infra |
| 2026-05-09 | DeepSeek V4 catalog: add v4-pro/v4-flash, fix V3-era pricing drift, schedule legacy alias deprecation | router intelligence catalog |
| 2026-05-09 | P0a — Agent budget enforcement (canonicalize agentLimits keys; closes Computer Protocol #244) | api security |
| 2026-05-09 | P0b — mTLS recovery hints + ECS misconfiguration warning (closes Computer Protocol #245 partial) | api security |
| 2026-05-09 | P1 #246 — forced tool_choice excludes reasoning models; reasoning-token cost safety net | router auto-selector |
| 2026-05-09 | Static-asset bundling — closes /openapi.json 503 + /.well-known/agents.json 404 + brk_ docs typo | api build docs |
| 2026-05-09 | Trust Envelope (A-1) — JOSE-aligned EdDSA schema, signing, audit-only middleware | security api |
| 2026-05-09 | TS-2 #248 — align audit, replays, streaming telemetry, feedback schema, cost center attribution | api completions telemetry |
| 2026-05-08 | Shadow comparator crosses statistical significance — Thompson +4.94% reward over static_price, p=0.002 | router intelligence comparator |
| 2026-05-08 | Chaos drill 05 (CAF cert revoke) executed against production — sub-ms pub/sub propagation | security caf chaos-drills |
| 2026-05-08 | ECS deploy workflow now pins all 8 provider API keys (closes silent-disable class) | infra ecs ci-cd ingestors |
| 2026-05-08 | Catalog ingestor coverage 7/7 → 8/8 — Groq closes the last drift gap | router intelligence ingestors |
| 2026-05-08 | Fix: model-performance persist no longer aborts with ON CONFLICT DO UPDATE error | router intelligence model-performance |
| 2026-05-08 | Catalog ingestor coverage 5/7 → 7/7 — Perplexity + Moonshot complete the set | router intelligence ingestors |
| 2026-05-08 | Savings ledger avgValidity wired end-to-end (was hardcoded 0 across all paths) | router intelligence savings-ledger quality-events |
| 2026-05-07 | Auth middleware: fold tenant status into Redis key cache + hoist dynamic import | api auth redis |
| 2026-05-07 | Budget middleware: 30s tenant-limits cache + Redis pipeline collapse | api budget redis |
| 2026-05-07 | Cache hygiene: bound 3 unbounded caches + O(1) LRU eviction | api security router cache infra |
| 2026-05-07 | Circuit breaker preserves forcedOpen on importFromRedis — closes cold-start breaker drift | router circuit-breaker |
| 2026-05-07 | Community-tier: collapse 5 sequential Redis ops into single pipeline | api community-tier redis |
| 2026-05-07 | Removed 1,501 LOC of orphaned channel-routing and stub modules | routing infra gateway cli |
| 2026-05-07 | Governance endpoints: cap unbounded ?limit on lineage and sampling-audit | security api governance |
| 2026-05-07 | Middleware: parallelize/pipeline 2 Redis hot paths | api middleware redis agent-reputation |
| 2026-05-07 | Misc: drop JSON.stringify token estimate + crypto.randomBytes for slug suffix | api community-tier provisioning |
| 2026-05-07 | Model performance: batch persistToDb into single multi-row INSERT | router intelligence persistence |
| 2026-05-07 | Advisory-mode mTLS wiring on agent-mesh write routes | security caf mtls agent-mesh |
| 2026-05-07 | Redis failure-mode coverage in 3 middleware sites — class-of-bugs fix | api middleware redis |
| 2026-05-07 | Semantic cache: partition entries by (tenantId + model) for O(partition) lookup | router cache intelligence |
| 2026-05-07 | Slack webhook: rate-limit BEFORE HMAC verification | security api admin |
| 2026-05-07 | Workspace agentId scoping + role-marker stripping in injected files | security workspace agents context-injection |
| 2026-05-06 | Circuit-breaker reset() now broadcasts to peers — kills the cross-task force-open feedback loop | router circuit-breaker observability sentinel |
| 2026-05-06 | Forensics panel 500 — listForTenant unwrapped pg.QueryResult.rows | api caf observability dashboard |
| 2026-05-06 | Sentinel recovery probe now bypasses model-list filter — closes the auto-recovery deadlock that PR #185 surfaced | router sentinel observability |
| 2026-05-06 | Cold-start timing wrapper dropped short-circuit Responses (5/14 5xx) | api middleware observability |
| 2026-05-05 | Layer 3 root-cause fix: DEFAULT_HEARTBEAT_EVERY was 600,000 minutes, not 600,000 ms | gateway infra |
| 2026-05-05 | Cut CloudWatch Logs ingestion ~99% by clamping setTimeout overflow at the runtime layer | gateway infra |
| 2026-04-30 | Explain narrator: BR uses its own router to write postmortems of its own routing decisions | router api sdk |
| 2026-04-30 | Provider doctor: BR explains its own circuit trips instead of waiting for paging | router/intelligence ops mcp |
| 2026-04-22 | BR Phase 1 — GET /v1/routing-stream (SSE push for routing decisions) | router sse api |
| 2026-04-22 | BR Tier 1 — god-mode audit table + CAF upsert-key cryptographic anchor | caf god-mode security-events |
| 2026-04-22 | BR Tier 2 — SDK mesh enrollment + hostname-collision handling | caf sdk-ts sdk-py mesh |
| 2026-04-21 | BR deploy pipeline Phase A — 18 min → 6 min target | ci-cd docker ecs vercel |
| 2026-04-21 | Add Kimi K2.6 to Moonshot catalog | model-router provider-catalog |
| 2026-04-20 | Marketing site reskin — complete (25/25 pages on new design) | marketing-site |
| 2026-04-20 | Marketing site reskin — Vite + Tailwind v4 foundation + homepage port | marketing-site |
| 2026-04-18 | ZTNA verifiability: build attestation, federated identity, reputation-coupled certs | caf attestation agent-identity continuous-verification sdk-ts sdk-py docs |
| 2026-04-16 | Self-soak test, streaming cascade stall detection, open-source prep | soak cascade ops cron monitoring governance |
| 2026-04-12 | Agent Session Loop — the 'run' verb ships with multi-provider per-turn routing | model-router agent-profiles mcp-gateway guardrails budget-enforcement session-store |
| 2026-04-10 | 58.3 → 78.5: Stochastic assessment remediation sprint | security testing operations intelligence documentation |
| 2026-04-10 | Sentinel: 60-second deep completion probes + CloudWatch alarms | router observability |
| 2026-04-07 | Agent Trap Defenses — 4 security fixes from DeepMind paper, 1 live bug patched | router memory security approvals |
| 2026-04-07 | API Security Audit — 5 fixes from 53-finding automated test suite | api router budget providers |
| 2026-04-07 | Intelligent Memory Architecture — 8 foundation items + 6 governance innovations, 149 tests | memory governance security compliance intelligence |
| 2026-04-01 | Wire all disconnected intelligence into production — 4 systems activated | ranking-engine savings-tracker phase-predictor semantic-cache |
| 2026-03-30 | Fix: 1200 false deprecation alerts eliminated | deprecation-detector model-intelligence-store provider-catalog |
| 2026-03-30 | Unified Agent Identity System — 6 attributes, 1 middleware, 4000 tokens | agent-context-injector heartbeat workspace reference soul skill memory |
| 2026-03-28 | Brainstorm CLI Integration P1-P4: Project Memory, Task Telemetry, Registry, Block Sync | memory api mcp sdk-ts sdk-py rbac |
| 2026-03-28 | Predictive Routing Intelligence — why THIS model for THIS request | |
| 2026-03-28 | Orchestration Trajectory Storage — the BrainstormLLM v2 training flywheel | trajectory-store api mcp sdk-ts sdk-py |
| 2026-03-27 | Agent Reputation Scoring — 6-Signal Behavioral Tier System | router api rbac |
| 2026-03-27 | Auto-Failover Explanation — per-request hop tracking with recovery headers | model-cascade circuit-breaker router |
| 2026-03-27 | Tier 1 Semantic Response Quality Scorer — 5 dimensions in <5ms | quality-scorer router |
| 2026-03-26 | Cost Prediction — Estimate Request Cost Before Execution | router api mcp sdk |
| 2026-03-26 | Explain Endpoint — One-Call Routing Decision Trace | router api mcp sdk |
| 2026-03-26 | Multi-Model Consensus: agreement-scored answer selection across 2-5 models | consensus-engine api-routes mcp-tools sdk-ts sdk-py |
| 2026-03-26 | SSE Streaming Heartbeat — zero-cost connection keep-alive | gateway streaming |
| 2026-03-25 | Agent Discoverability Foundation — Fix the Front Door | api mcp sdk-ts sdk-py intelligence |
| 2026-03-25 | Community API Key — Zero-Setup Onboarding for Brainstorm CLI | api middleware db |
| 2026-03-25 | Memory endpoints now accept API key auth with per-key isolation | memory auth |
| 2026-03-25 | Model Recommend: data-driven model selection advisory for agents | model-bandit cost-optimizer performance-tracker auto-selector |
| 2026-03-25 | Structured Error Recovery: Every error tells the agent what to do next | error-recovery router intelligence |
| 2026-03-24 | Delegated routing: Brainstorm CLI delegates model selection to BR's 13 intelligence systems | model-auto-selector model-bandit circuit-breaker ranking-engine performance-tracker |
| 2026-03-24 | Distributed Mesh Execute — DAG-Based Workflow Orchestration | api sdk-ts sdk-py mcp |
| 2026-03-24 | Experiment Framework — A/B Test Models with Real Tasks | api sdk-ts sdk-py |
| 2026-03-24 | Fleet Intelligence — Cross-Tenant Model Insights from All Brainstorm Instances | api sdk-ts sdk-py mcp |
| 2026-03-24 | Outcome feedback: every code compile/test/acceptance feeds Thompson sampling | performance-tracker model-bandit thompson-sampling |
| 2026-03-24 | Parallel model comparison: fan same prompt to 2-6 models, compare quality/cost/latency | model-router validity-scorer performance-tracker |
| 2026-03-24 | Project Config & Activity — Per-Repository Settings and Collaboration Awareness | api sdk-ts sdk-py mcp |
| 2026-03-24 | Real-Time Telemetry SSE — live model health, budget alerts, routing decisions | router circuit-breaker budget-forecaster observability |
| 2026-03-24 | Session State Persistence — Resume Brainstorm Sessions from Any Machine | api sdk-ts sdk-py |
| 2026-03-24 | Task Pattern Memory — Institutional Knowledge of What Approaches Work | memory api sdk-ts sdk-py mcp |
| 2026-03-23 | A2A Attack Detection — multi-hop injection chain tracking | security |
| 2026-03-23 | A2A Public Facade & Behavior Certificates | api security |
| 2026-03-23 | Automated Red Teaming — 30+ adversarial vectors, 5 categories | security |
| 2026-03-23 | Behavioral Fingerprinting & Bot Detection — 5 signal dimensions | security |
| 2026-03-23 | Canary / A-B Model Routing — traffic splits with winner detection | router |
| 2026-03-23 | Compliance Dashboard — SOC2, audit viewer, data residency, policies | dashboard compliance |
| 2026-03-23 | Custom Request Metadata — x-br-metadata header | api observability |
| 2026-03-23 | Self-Hosting Documentation | distribution |
| 2026-03-23 | Evaluation Dashboard — dataset browser, run launcher, comparison | dashboard evaluation |
| 2026-03-23 | Evaluation Runs & Scoring Engine — 6 scorers, statistical comparison | evaluation |
| 2026-03-23 | Evasion Technique Decoders — 7-stage normalization pipeline | security |
| 2026-03-23 | Free Tier & Plan System — free/pro/enterprise with enforcement | billing |
| 2026-03-23 | GenAI Entity Summaries — AI-powered 'Explain This' for dashboard entities | dashboard agents security observability |
| 2026-03-23 | Intent Detection Engine — 4 adversarial intent types with confidence scoring | security |
| 2026-03-23 | MCP Server Vetting — trust scoring, tool manifest scanning | security mcp |
| 2026-03-23 | Agent Mesh Graph — interactive topology visualization | dashboard agents security |
| 2026-03-23 | Model Self-Defense — security-aware system instructions per trust level | security |
| 2026-03-23 | OpenTelemetry Trace Export — Gen AI semantic conventions | observability |
| 2026-03-23 | OWASP LLM Top 10 Coverage — automated compliance check | security compliance |
| 2026-03-23 | Prometheus /metrics Endpoint — counters, histograms, gauges | observability |
| 2026-03-23 | Per-Request Provider Filtering — only, ignore, ZDR | router |
| 2026-03-23 | Quick-Start Wizard — 4-step onboarding with provider validation and security presets | dashboard onboarding |
| 2026-03-23 | Shadow AI Cloud Connector — multi-source discovery with persistent baselines | security observability |
| 2026-03-23 | Shadow AI Discovery — detect unknown models, providers, IP anomalies | security |
| 2026-03-23 | SOC2 Audit Trail — hash chain, retention, CSV/JSON/CEF export | security compliance |
| 2026-03-23 | Virtual Model Variants — brainstormrouter/fast|cheap|quality | router |
| 2026-03-22 | Audio API — TTS + transcription with provider routing and cost tracking | api router |
| 2026-03-22 | Evaluation Datasets — DB-backed test case management | evaluation database |
| 2026-03-22 | Moderations API — POST /v1/moderations with BR guardrail layer | api security |
| 2026-03-22 | Dashboard Prompt Editor — visual template management with version history | dashboard prompts |
| 2026-03-22 | Prompt Evaluation Engine — dataset-based scoring with 4 strategies | prompts evaluation |
| 2026-03-21 | Agent-as-Employee UX — trust shields, timelines, comparison, prominent kill switch | dashboard agents |
| 2026-03-21 | Analytics Charts — time range selector, sparklines, inline decision trace, circuit badges | dashboard observability |
| 2026-03-21 | Dark Security Control Plane Theme — CrowdStrike-grade dashboard aesthetic | dashboard |
| 2026-03-21 | Image Generation API — POST /v1/images/generations with provider routing and cost tracking | api router |
| 2026-03-21 | Security Dashboard UX — threat banner, posture score, pipeline diagrams, governance urgency | dashboard security |
| 2026-03-20 | Agent Governance + Security Controls — First-Class Dashboard Governance | dashboard agents security killswitch guardian policy-engine |
| 2026-03-20 | Cost Attribution Insights -- Usage by Key and Agent | usage-queries auth-routes sdk-ts sdk-py dashboard |
| 2026-03-20 | Dashboard Design System + Real-Time SSE Pipeline | dashboard observability api sdk |
| 2026-03-20 | Dashboard Polish — Component Library + Cross-Page Consistency | dashboard |
| 2026-03-20 | Decision Trace — Full Routing Pipeline Visibility per Request | |
| 2026-03-20 | Routing Brain Dashboard + Home Page Live — 6 Intelligence Sections, 3 New Endpoints | |
| 2026-03-19 | Intelligence Phase 2: Benchmark Priors, Task Classification, Cross-Tenant Ranking | ranking-engine model-performance model-auto-selector benchmark-ingestor |
| 2026-03-19 | Intelligence Phase 3: Live Ingestors, Deprecation Trends, Tenant Availability | model-intelligence-engine deprecation-detector model-auto-selector model-router |
| 2026-03-19 | Intelligence Phase 4 — Cost-Quality Frontier API + Public Site Visualizations | intelligence cost-optimizer ranking-engine sdk-ts sdk-py site |
| 2026-03-18 | Crypto Agility Layer + Hybrid PQC TLS Foundation | crypto-agility tls pqc |
| 2026-03-18 | Dead Code Removal + Config Drift Wiring | cleanup config-drift |
| 2026-03-18 | Operation Deep Audit — 6 Intelligence Gaps Fixed, System Score 69→81 | |
| 2026-03-18 | MCP Capabilities Endpoint + SDK Methods | mcp sdk |
| 2026-03-18 | Memory Extraction + Prompt Rewrite: Pipeline Activation | memory-extraction prompt-rewrite |
| 2026-03-18 | PQC Docs + Security Page: Post-Quantum Positioning | pqc docs marketing |
| 2026-03-18 | Streaming Semantic Cache: Cache Hits for ~50% of Traffic | semantic-cache streaming model-router |
| 2026-03-17 | AI-Native Surface: 65 MCP tools, ambient intelligence headers, and GET /v1/discovery | router auto-selector mni discovery error-recovery |
| 2026-03-17 | Audit HMAC: Auto-Load Signing Key from Secrets Manager | audit-signer |
| 2026-03-17 | Infrastructure Hardening: Cold-Start Elimination, SSE Sanitization, Atomic Budget | gateway guardian budget streaming |
| 2026-03-17 | Redis CAS: Cross-Process Audit Chain Atomicity | audit-chain anomaly-engine mcp-governance |
| 2026-03-17 | RMM Durable Memory: EFS + Postgres Fallback | rmm memory paths |
| 2026-03-17 | Wire SavingsTracker + Extend Anomaly Scoring to API Key Callers | savings-tracker anomaly-engine model-router |
| 2026-03-17 | Vanguard Sprint: SLM anomaly triage wired into mesh scoring | anomaly-engine model-router caf |
| 2026-03-16 | Brain Intelligence Plugin — Live model intelligence injected into every API completion | plugins router completions mcp-bridge |
| 2026-03-15 | Anthropic Messages API facade — native Claude SDK compatibility | api router identity-bridge |
| 2026-03-15 | MNI v2: Agent-Native Runtime Interface — Self-Awareness, Self-Healing Errors, Transparent Health | |
| 2026-03-14 | Guardrails V2: 8 Composable Checks, Pipeline Builder, 47 Tests | guardrails pipeline-executor check-registry |
| 2026-03-14 | Identity Bridge: Machine-to-Dashboard Account Linking | identity auth provisioning dashboard |
| 2026-03-13 | Digital Worker Manifest V2 — Workday for AI Agents | agent-profile-store agent-bootstrap auth-bridge dashboard |
| 2026-03-13 | Rich 403s + Permission Introspection — Self-Healing RBAC for Agents | rbac api-key-management mcp sdk-ts sdk-py |
| 2026-03-12 | Phase 012: Depth Audit — Crypto, Forensics, Connectors, MCP, Guardrails, Approvals, Memory, Cache | cryptography forensics connectors mcp guardrails approvals memory semantic-cache |
| 2026-03-11 | Budget Intelligence: Forecaster, Agent Budgets, and Alerts | budget-forecaster agent-budget-manager budget-alerts |
| 2026-03-11 | Tool Firewall: Mask, Substitute, Downgrade Actions | |
| 2026-03-11 | Phase 011: The Obsidian Catch-Up — 15/15 Spec Areas SHIPPED | |
| 2026-03-11 | Scoped Kill Switches + Policy Dry-Run: Granular Emergency Controls | kill-switch policy-engine governance |
| 2026-03-10 | Agent Resource Management Convergence — 17 endpoints, 15 SDK methods, 8 MCP tools unified | agent-profiles agent-bootstrap agent-delegation agent-lifecycle |
| 2026-03-10 | Budget Controls & Kill Switch — Unified Governance MVP | budget-management kill-switch governance |
| 2026-03-10 | Forensics & Replay: Full SDK + MCP + Dashboard Coverage | routing-replay forensic-snapshots mcp-tools sdk-ts sdk-py dashboard |
| 2026-03-10 | MCP Control Plane — 43 Tools, Secretless Access, Full Operator Governance | mcp-server mcp-gateway governance dashboard |
| 2026-03-10 | Memory Runtime Foundation — 5 MCP tools, full SDK parity, search UI | memory mcp sdk dashboard |
| 2026-03-10 | Runtime Security Convergence — 4-Stage Pipeline with Guardrail MCP Tools | security guardrails mcp dashboard |
| 2026-03-10 | Enterprise Documentation Consolidation — 5 Reference Pages, OAuth SDK Parity | admin-api scim oauth kms tenant-settings sdk-ts sdk-py mcp |
| 2026-03-09 | SDK Parity Hardening — 4 resources, 15 methods, 8 MCP tools | config insights capacity aliases |
| 2026-03-08 | MCP Gateway Fix Sprint — 4 Protocol-Breaking Bugs Eliminated | mcp-gateway mcp-governor mcp-client agent-identity sdk-ts sdk-py |
| 2026-03-08 | Provider Health Alerts — Automated Outage Detection Across 247 Endpoints | endpoint-sentinel model-auto-discovery platform-event-bus notifications |
| 2026-03-07 | Graduated Trust Degradation — Continuous Agent Governance | anomaly-engine model-router agent-profiles caf-mesh anomaly-baselines |
| 2026-03-05 | Admin Intelligence — Agent-First Business Management | platform-event-bus beta-gate admin-api invite-codes slack-interactive |
| 2026-03-05 | Agent Bootstrap — Zero-Human Onboarding in One API Call | agent-profiles agent-budget-manager caf-mtls security-events |
| 2026-03-04 | ARM: Agent Resource Management — The Workday for AI Agents | agent-profiles agent-budget-manager virtual-keys anomaly-kill-switch |
| 2026-02-25 | Streaming Tool Firewall: 7-Check Pipeline, Zero Extra LLM Calls | tool-call-firewall streaming-guardrails |
| 2026-02-22 | Virtual Key Vault: BYOK with AES-256-GCM Encryption and Zero-Downtime Rotation | virtual-keys provider-crud |
| 2026-02-20 | CAF: Cryptographic Agent Framework with 5-Minute Certificates | caf-mtls caf-mesh spiffe anomaly-kill-switch |
| 2026-02-19 | Relational Memory Manager: 4-Block Structured Context with Relevance Scoring | relational-memory memory-api |
| 2026-02-18 | Guardian Intelligence: Sub-5ms Cost Prediction on Every Request | guardian cost-prediction usage-recording velocity-tracking |
| 2026-02-16 | Universal Adapter Matrix: 5 Providers, 1 Interface, BigInt Precision | model-executor adapter-registry |
| 2026-02-15 | Thompson Sampling: Principled Model Selection Across 247 Endpoints | model-bandit model-performance-tracker |
| 2026-03-26-budget-warnings-inband | |
| 2026-03-26-context-window-utilization | |
| 2026-03-26-degradation-ladder | |
| 2026-03-26-guardrail-transparency | |
| 2026-03-26-model-deprecation-warnings | |
| 2026-03-26-per-model-capabilities | |
| 2026-03-26-request-pattern-fingerprinting | |
| 2026-03-26-session-context-hints | |
| 2026-03-26-tool-call-success-tracking | |
| 2026-05-08-anthropic-catalog-ingestor | |
| 2026-05-08-cache-prefix-backfill-v51 | |
| 2026-05-08-cache-prefix-nesting | |
| 2026-05-08-capacity-bump-and-mtls-strict | |
| 2026-05-08-doc-state-drift-gate | |
| 2026-05-08-drill-02-redis-auth-failure | |
| 2026-05-08-drill-03-provider-key-rotation | |
| 2026-05-08-drill-04-task-stop-mid-stream | |
| 2026-05-08-e2e-suite-green | |
| 2026-05-08-fix-job-queue-redis-duplicate-error-handlers | |
| 2026-05-08-load-test-r20 | |
| 2026-05-08-openai-catalog-ingestor | |
| 2026-05-08-sandbox-savings-pollution | |
| 2026-05-08-three-more-ingestors | |
| 2026-05-10-279-auto-selector-gates-on-sentinel | |
| 2026-05-10-docs-trust-envelope-x2 | |
| 2026-05-10-f1-disclosure-committee-hitl-gate | |
| 2026-05-10-fix-issues-272-273-274 | |
| 2026-05-10-gauntlet-skeleton-alpha-5 | |
| 2026-05-10-x2-event-taxonomy-expansion | |
| 2026-05-10-x2-pii-emission | |
| 2026-05-11-281-schema-v0.2-extensions | |
| 2026-05-11-alpha5-v0.3-migration | |
| 2026-05-11-envelope-precedence-docs-drift | |
| 2026-05-11-envelope-stress-tests | |
| 2026-05-11-envelope-synth-defensive-clamps | |
| 2026-05-20-bandit-reward-v2 | |
| 2026-05-20-invariant-test-layer | |
| 2026-05-20-openclaw-phase-7-verification | |
| 2026-05-20-openclaw-stress-in-ci | |
| 2026-05-20-phase-8-1-2-contract-detect-cache-skip | |
| 2026-05-20-phase-8-3-contract-verifier | |
| 2026-05-20-phase-8-4-verifier-reward-hook | |
| 2026-05-20-waf-cmdi-backtick-narrow | |
| 2026-05-21-phase-8-5-cache-store-verifier | |
| 2026-06-10-inc1-scoped-agent-grants | |
| 2026-06-11-inc1.5-audit-chain-crash-safety | |
| 2026-06-11-inc2-unified-scope-enforcement | |
| 2026-06-12-inc3-capability-grants-delegation | |
| 2026-06-12-inc4-provenance-traceparent | |
| 2026-06-12-trackb-governance-dashboard-design | |