AGENTIX · CASE FILE / 01 SAAS PLATFORM · TECHAPPFORCE / 2026
§ Case file · production system

TAFI

The AI co-builder for TechAppForce

Describe an app in plain English. A multi-agent pipeline drafts the requirements, designs the data model, plans the build as a dependency graph, and registers every entity, screen, query and workflow on the live low-code platform, pausing for your approval at four gates and verifying every output so nothing is fabricated. A certified run built a 27-step leave-management app in ~12 minutes.

0 min to build a 27-step app · certified run
0/279 eval stories · E0–E12
0 QA score · 3 clean runs
0 ~commits across 5 repos
FIG.00, plain-English prompt → registered TechAppForce app · illustrative diagram, not a screenshot
PROMPT · plain English

“Build a leave-management app, employees request time off, a manager approves or rejects, and HR sees a balance per person.”

⛓ Requirements ⛓ Blueprint ⛓ Build ⛓ Release
REGISTERED ON TECHAPPFORCE
  • 4 entities · LeaveRequest, Employee, Balance, Policy
  • 5 screens · request, approve, balance, admin, list
  • 2 workflows · approval + balance-decrement
  • 27 build steps · 96 platform writes
QA 100contract-verified~12 min

Values shown are the canonical figures from the certified 27-step run, 96 platform writes, QA 100, 3 consecutive clean runs.

§ The product, live

Not a mockup.
The shipping app, running.

The same pipeline inside the live TechAppForce builder, replayed in the order a build actually runs: describe an app in plain English, watch TAFI plan and build it step by step, and find every build tracked in one workspace.

TAFI · Describe: Describe an app in plain English — or upload a requirements doc.
TAFI · Building: TAFI builds step by step through a live plan — you approve at each stage.
TAFI · Shipped: Every build, a registered TechAppForce app — tracked in one workspace.

Describe an app in plain English — or upload a requirements doc.

FIG · TAFI live replay · real captures from app.techappforce.com, sequenced describe → build → shipped
The build

One description in.
A working application out.

Three milestones, each gated by a human approval. The whole pipeline is wrapped by a typed MCP gateway and grounded by a cited knowledge base, so nothing reaches the platform unverified. This is the pattern behind our agentic pipelines practice.

FIG.01, TAFI agentic build pipeline · plain English in, registered TAF app out · ⛓ = human approval gate
TAFI agentic build pipeline: a plain-English app description passes Requirements, Blueprint+Plan and Build milestones, each behind a human approval gate, making 96 platform writes through a typed FastMCP gateway, verified by contract checks, to a registered TechAppForce app. 279/279 stories, QA 100, 3 clean runs. Every enum cited from C# source (zero hallucination).
§01, The situation

Expert developer time,
measured in hours.

Building an application on TechAppForce takes specialist developer time measured in hours. A developer manually coordinates dozens of API writes across six subsystems in exact dependency order, with no tolerance for sequencing errors.

Mistakes are silent, phantom IDs and broken field references surface only at QA, forcing rework that compounds the original overhead. The cognitive load resists delegation: it needs platform expertise that documentation can't transfer, and that expertise becomes the bottleneck as usage grows.

Without TAFI

A developer hand-coordinates API writes across 6 subsystems. Hours per app. Phantom IDs surface at QA. Platform expertise is the bottleneck.

With TAFI

Type a plain-English description. TAFI generates the full application, 96 API calls, correct dependency order, propagation verified. No manual steps.

§02, What we built

Six parts, each killing
a class of failure.

Every component exists to eliminate a specific failure the previous approach hit at scale, not for novelty.

Multi-Agent Pipeline (CrewAI)

A planner decomposes the request into a dependency graph; an executor walks it with level-based parallelism and per-step isolation; specialist worker crews run each step. Failures trigger bounded replanning, not a crash.

15 KB-Driven Skills

One specialist "designer" skill per artifact class, each a YAML manifest with a manager+reviewer dual-agent pattern and an output contract. Add or change a skill by editing a doc, not code.

MCP Integration Gateway

All platform writes route through the FastMCP gateway: reads over REST, writes over RabbitMQ/MassTransit. An idempotent lookup→normalize→publish→poll composer guarantees exactly-once registration, re-runs never duplicate.

Contract Verifier, 8 Rules

Hard-rejects fabricated UUIDs and "claimed success with no tool call," and cross-checks each agent's claimed calls against the gateway's real count. The fabricated-data class is killed structurally, not prompted away.

Knowledge Base, Mem0 + BM25 + Graph

Vector retrieval raced against a BM25 keyword fallback, plus deterministic link-graph expansion. Every enum value is cited from C# source at file:line, a grounded, zero-hallucination reference for every field decision.

Provider-Agnostic Routing (LiteLLM)

Five model tiers with fallback chains, reasoning-tier escalation by entity count, a hard $5/session budget cap and a dry-run mode. No vendor lock-in; cost is bounded before a build starts.

§03, Guarantees

Six promises most code-
generation can't make.

A model that writes plausible code can't tell you it didn't invent an enum, won't duplicate a record on retry, or will refuse to ship unverified. TAFI can, each guarantee is enforced by structure, not by a prompt.

01

Zero-hallucination enums

Field types, query operators, blueprint statuses and screen types are not free-text the model can invent, they are read from the Scopes/ layer, 10 code-verified directories where every value is cited from C# source.

→ Cited from source · file:line · 10 scope dirs

02

Idempotent platform writes

Every registration runs lookup → normalize → publish → poll, then waits until the write is confirmed visible on the TAF REST API. Re-running a build repairs state; it never duplicates a record.

→ Propagation poll 20×5s · safe to re-run

03

Strict dependency ordering

The planner emits a DAG; the executor walks it level by level. A skill never references an ID an earlier stage has not yet committed, so phantom IDs cannot enter the graph in the first place.

→ 8 dependency stages · level-based walk

04

Eval-gated shipping

Nothing ships on a green eval alone. A build must pass the deterministic harness (Score = 100 − Σ penalties) and then clear three consecutive clean runs before it is certified.

→ QA 100 · 3 clean runs required

05

Contract-verified output

The eval scored Equipment Tracker 100 yet platform verification still caught five failure classes, dead-lettered queries, empty screen JSON, consumer-less bus commands, missing nav columns, no audit trail. All five are now checked against live state.

→ 5 failure classes the score was blind to

06

Platform-agnostic by construction

Zero TAF-specific strings live in the production runtime (E13.F6). Supporting a new low-code platform is a YAML config plus KB documents, not a new code path through the pipeline.

→ New platform = YAML + KB · 0 new code

§ Governed & verified by design

The reason an enterprise can trust an autonomous builder: every output is gated, checked, and idempotent, not "trust the LLM."

4 approval gates

Requirements · Blueprint · Implementation · Release. Approve, request a change in plain English, or rewind, never a black box.

Verified, not hoped

Eight contract rules reject fabrication; agent claims are reconciled against real gateway calls. "Done" means provably done.

Idempotent by design

Lookup-before-create plus poll-for-visibility means a re-run repairs, never duplicates. Exactly-once registration on a live platform.

§04, Our definition of done

A passing eval is
not the finish line.

Most teams ship when the eval goes green. We don't. Every build clears two independent gates, and only ships when both agree.

GATE 1 Eval harness Every story in the suite passes · QA 0–100
GATE 2 Platform verification Checked against real, live platform state
RESULT Ships Only when both gates agree
Proof · TAFI epic E13

On TAFI, the eval returned QA 100, every story passed. Gate 2 still caught five issues the eval was blind to, all invisible to a score:

  • Queries dead-lettered in the message bus
  • Screen-container JSON rendering empty
  • Workflow bus commands with no TAF consumer registered
  • Navigation records missing required columns
  • No audit trail on artifact generation

All five fixed, then three consecutive clean runs. The score only became trustworthy once it matched verified platform state, which is the discipline behind every eval number Agentix publishes. We run the same eval / red-team / regression discipline on shipped AI as AI Assurance, and document the method in how we work.

§05, The taxonomy

15 skills, resolved into
8 dependency stages.

Each skill is one specialist designer, a YAML manifest with a manager+reviewer dual-agent pattern and an output contract. The planner sorts them into a DAG so every stage runs only once its inputs exist. Two meta-skills sit outside the enum: they run the pipeline, not a step in it.

STAGE 0 1 skill

Analyst

analyst

Plain-English prompt → RequirementsDoc. No platform writes, pure intake.

STAGE 1 1 skill

Blueprint Architect

blueprint_architect

Build order, entity relationships, has_workflow flags. Artifact-only; the spine every later stage reads.

STAGE 2 2 skills

Entities

blueprint_registrardb_designer

MSSQL DDL + AppObject registration; workflow blueprint entities registered first.

STAGE 3 1 skill

Queries

query_designer

Data Source Queries built against the AppObjects registered one stage up.

STAGE 4 3 skills

Screens & Components

component_designerscreen_designerscreen_validator

form.io component selection, schema + DSQ binding, then a read-only JSON integrity pass.

STAGE 5 3 skills

Parallel Execution

extension_designertheme_designerworkflow_designer

Three independent skills run concurrently, no inter-dependency inside the stage.

STAGE 6 1 skill

Navigation

navigation_designer

Modules, categories and menu items wired to the registered screens.

STAGE 7 1 skill

Routing

routing_designer

Site map + routing config, the last layer before the app is navigable.

Pipeline-invoked · outside the planner enum
planner Reads the Blueprint, emits the ordered DAG
qa DB assertions + LLM quality judge across every artifact
§06, Results

What changed,
measured.

Not projections, the measured delta between the manual baseline and the shipped system.

Specialist-days ~12 min

to build a 27-step leave-management app · certified run, 96 platform writes, QA 100

Code deleted, not added
MCP bridge1,038 → 200 lines
−81%
Agent-layer code (V3 → V4)~21,200 lines removed
−76%
Stories delivered 0279 / 279
QA score N/A100 · 3 clean runs
Fabricated-ID bugs every sessioneliminated
Platform support 1 (TAF)any · YAML + KB
§07, Engineering depth

~1,615 commits.
Five repositories.

Sustained, parallel delivery, not a vanity number. The 1,428 commits tracked across the five repos below were authored by Sagar Maheta, Vaishali Kochar, Mohit Jagatiya and Bhavik Thakkar, with AI collaborator Sammy on the knowledge base.

  • taf_tafi Core pipeline engine 951
  • superproject Root ecosystem coordinator 116
  • taf_knowledgebase 21 Pattern docs · Scopes/ · surface index 162
  • taf_mcp MCP gateway · 11 tools 127
  • taf_tafifrontend Angular 19 · 32 components 72
1,038 → 200LOC MCP bridge, after auto-discovery 81% deleted, tools discovered at startup, none hardcoded
2,500 → 0LOC Reconciler, replaced by 11 smart tools The fabricated-ID class eliminated structurally, not patched
21docs Pattern documents ingested into Mem0 The only KB layer agents read at runtime (Qdrant)
§ Stack

Open-source.
Self-hosted. Owned.

Every layer owned by the client from day one, no vendor lock-in, no ongoing platform fees.

CrewAIFastAPIFastMCPMassTransitRabbitMQQdrantMem0Angular 19SSEPydanticLiteLLMLangfuse
Start a project

Want a system like
TAFI for your platform?

Tell us what you're building. We'll tell you how we'd architect it, what it would take, and the eval it would have to pass before it ships.

← Back to all systems

AGENTIX TECH · case file 01 · TAFI · production