PCProvider ComplianceAgent workflow and memory
GitHub lane 3 · 40 verified repositories · 12 deep profiles

Stateful agents are workflow components—not the case database.

LangGraph is the closest single agent framework. Temporal is the high-assurance execution reference. Supabase + LangGraph + Langfuse is the strongest small AI composition, but the first proof should begin with direct state transitions and a typed case authority.

Memory is derived context, not truth.

Applicant facts, approved decisions and source versions live in the case database. Agent memory may hold consultant preferences or reviewed lessons, but it must carry tenant, case, source, sensitivity, approval and supersession metadata.

First-build ruleImplement intake → propose → ask → approve → validate as explicit application state. Add an agent graph only when real cases prove the branching/resume benefit.

Eight retrieval namespaces

regulatory_authorityinstructionsmaster_templatespolicy_memorycase_factscase_historyevaluation_fixturesexternal_research

Tenant and case filters are applied before retrieval. Similarity ranking never expands the authorization boundary.

Twelve strongest workflow and memory references

Closest agent framework

LangGraph

Typed graph state flows through nodes/edges. Checkpoints persist threads; interrupts pause and later resume with human input.

Use
Branching draft/ask/approve/validate flows.
Missing
Authorization, case semantics and business audit.
Repository ↗
High-assurance execution

Temporal

Deterministic workflow code replays from event history; activities perform side effects; signals and timers resume long waits.

Use
Reliable multi-day waits and failure recovery.
Burden
Service operations and deterministic worker discipline.
Python SDK ↗
Durable services

Restate

Journal-backed service handlers, virtual objects, durable promises and timers survive process failure.

Use
Stateful callbacks and exactly-once-style coordination.
Risk
Licence metadata and integration burden.
Repository ↗
TypeScript waitpoints

Trigger.dev

Deployed tasks use run IDs, retries, queues, concurrency controls and programmable human waitpoints.

Use
Fast TS long-task delivery.
Missing
Case/evidence semantics and RLS.
Repository ↗
Event-driven steps

Inngest

Event-triggered functions split work into durable steps with retries, sleeps and cancellation.

Use
Lightweight integration workflow.
Risk
Licence and hosted-data boundaries.
Repository ↗
Derived memory only

Mem0

Processes messages into candidate memories, deduplicates/updates them and retrieves them later.

Use
Consultant preferences and reviewed lessons.
Prohibited
Authoritative applicant facts.
Repository ↗
Context service

Zep

Stores conversation and structured context for later retrieval and assembly.

Use
De-identified context experiments.
Missing
Fact approval and regulatory authority.
Repository ↗
Retrieval index

Qdrant

Vectors plus payload metadata support filtered similarity search across tenant, case, namespace and edition.

Use
Derived chunk index after exact filtering.
Boundary
Never canonical document/evidence storage.
Repository ↗
Relationship authorization

OpenFGA

Zanzibar-style relationship tuples answer whether an actor may perform an action on a case or artifact.

Use
Complex consultant/client/reviewer sharing.
Trigger
Only when RLS becomes intractable.
Repository ↗
Deterministic policy

Open Policy Agent

Rego evaluates structured input and returns policy decisions separately from enforcement.

Use
Category, expiry, transition and submission gates.
Trigger
Large versioned rule catalogue.
Repository ↗
LLM observability

Langfuse

Receives traces for model, retrieval and agent calls; versions prompts/datasets and compares runs.

Use
Which prompt/model/source produced a draft.
Boundary
Not the legal/business audit.
Repository ↗
Fixture evaluation

Promptfoo

Declarative tests run prompts/providers against assertions, metrics and red-team cases locally or in CI.

Use
Missing-fact, non-invention, category and injection tests.
Fit
First optional AI addition.
Repository ↗

Closest combinations

CompositionBest forDecision
Supabase + LangGraph + LangfuseCase data plus graph interrupts and tracesBalanced pilot after baseline
Supabase + Temporal + LangfuseHigh-assurance durable executionLater-scale option
Supabase + Trigger.dev + PromptfooFast TypeScript pilot and regressionAlternative delivery path
LangGraph + OpenFGA + Qdrant + PromptfooRetrieval-heavy agent systemStill missing canonical case DB

What remains custom

Case aggregate, source/edition manifests, claim/evidence lineage, consultant/client gap ownership, approval/declaration records, regulatory validators, append-only business audit, PII lifecycle, tenant-isolation tests and successful/incomplete fixtures.

Read the minimum-repo decision