From Black Field to Blueprint

Key Challenges

1. Lacking Supply Code: legacy understanding is already complicated if you
   have supply code and an SME (in some type) to place every part collectively. When the
   supply code is lacking and there are not any consultants it’s a good larger problem.
   What’s left are some compiled binaries. These will not be the current binaries that
   are straightforward to decompile on account of wealthy metadata (like .NET assemblies or JARs), these
   are even older binaries: the type that you simply may see in outdated home windows XP underneath
   C:Home windowssystem32. Even when the database is accessible, it doesn’t inform
   the entire story. Saved procedures and triggers encode many years of accrued
   enterprise guidelines. Schema displays compromises made based mostly on context unknown.
2. Outdated Infrastructure: OS and DB reached finish of life, long gone its
   LTS. Utility has been in a frozen state within the type of VM resulting in
   vital threat to not solely enterprise continuity, additionally considerably rising
   safety vulnerability, non compliance and threat legal responsibility.
3. Institutional Data Misplaced: whereas hundreds of finish customers are
   repeatedly utilizing the system, there’s hardly any enterprise data out there
   past the occasional assist actions. The stay system is one of the best supply of
   data. The one dependable view of performance is what customers see on display.
   However the UI captures solely the “final mile” of execution. Behind every display lies a
   tangled internet of logic deeply built-in to a number of different core programs. This can be a
   widespread problem, and this technique was no exception, having a historical past of a number of
   failed makes an attempt to modernize.

Our Aim

The target is to create a wealthy, complete purposeful specification
of the legacy system without having its authentic code, however with excessive
confidence. This specification then serves because the blueprint for constructing a
trendy alternative utility from a clear slate.

• Perceive total image of the system boundary and the mixing
  patterns
• Construct detailed understanding of every purposeful space
• Establish the widespread and distinctive eventualities

To make sense of a black-box system, we would have liked a structured option to pull
collectively fragments from totally different sources. Our precept was easy: don’t
attempt to recuperate the code — reconstruct the purposeful intent.

Our Multi Lens Method

It was a 3 tier structure – Internet Tier (ASP), App Tier (DLL) and
Persistence (SQL). This structure sample gave us a leap begin even with out
supply repo. We extracted ASP recordsdata and DB schema, saved procedures from the
manufacturing system. For App Tier we solely have the native binaries. With all
this data out there, we deliberate to create a semi-structured
description of utility habits in pure language for the enterprise
customers to validate their understanding and expectations and use the validated
purposeful spec to do accelerated ahead engineering. For the semi-structured
description, our method had broadly two elements

1. Utilizing AI to attach dots throughout totally different knowledge sources
2. AI assisted binary Archaeology to uncover the hidden performance from
   the native DLL recordsdata

UI Layer Reconstruction

Shopping the prevailing stay utility and screenshots, we recognized the
UI components. Utilizing the ASP and JS content material the dynamic behaviour related
with the UI component may very well be added. This gave us a UI spec like under:

What we regarded for: validation guidelines, navigation paths, hidden fields. One
of the important thing challenges we confronted from the early stage was hallucination, each
step we added an in depth lineage to make sure that we cross verify and make sure. In
the above instance we had the lineage of the place it comes from. Following this
sample, for each key data we added the lineage together with the
context. Right here the LLM actually sped up the summarizing of huge numbers of
display definitions and consolidating logic from ASP and JS sources with the
already recognized UI layouts and subject descriptions that might in any other case take
weeks to create and consolidate.

Discovery with Change Information Seize (CDC)

We deliberate to make use of Change Information Seize (CDC) to hint how UI actions mapped
to database exercise, retrieving change logs from MCP servers to trace the
workflows. Setting constraints meant CDC may solely be enabled partially,
limiting the breadth of captured knowledge.

Different potential sources—comparable to front-end/back-end community site visitors,
filesystem adjustments, extra persistence layers, and even debugging
breakpoints—stay viable choices for finer-grained discovery. Even with
partial CDC, the insights proved helpful in linking UI habits to underlying
knowledge adjustments and enriching the system blueprint.

Server Logic Inferance

We then added extra context by supplying
typelibs that had been extracted from the native binaries, and saved procedures,
and schema extracted from the database. At this level with details about
format, presentation logic, and DB adjustments, the server logic will be inferred,
which saved procedures are possible referred to as, and which tables are concerned for
most strategies and interfaces outlined within the native binaries. This course of leads
to an Inferred Server Logic Spec. LLM helped in proposing possible relationships
between App tier code and procedures / tables, which we then validated by means of
noticed knowledge flows.

Prior Makes an attempt

Earlier than we arrived at this method, we tried

• brute-force technique: Added all meeting features right into a workspace, and used
  the LLM agent to make it humanly readable pseudocode. Confronted a number of challenges
  with this. Ran out of the 1 million context window as LLM tried to finally
  load all features on account of dependencies (references it encountered e.g. operate
  calls, and different features referencing present one)
• Cut up the set of features into a number of batches, a file every with 100s of
  features, after which use LLM to investigate every batch in isolation. We confronted so much
  of hallucination points, and file measurement points whereas streaming to mannequin. A number of
  features had been transformed meaningfully however plenty of different features did not make
  any-sense in any respect, all seemed like related capabilities, on cross checking we
  realised the hallucination impact.
• The subsequent try was to transform the features one after the other, to
  guarantee LLM is supplied with a contemporary slender window of context to restrict
  hallucination. We confronted a number of challenges (API utilization restrict, fee
  limits) – We could not confirm what LLM translation of enterprise logic
  was proper or mistaken. Then we could not join the dots between these
  features. Attention-grabbing observe, we even discovered some C++ STDLIB features
  like
  std::vector::insert
  on this method. We discovered so much had been truly unwind features purely
  used to name destructors when exception occurs (stack
  unwinding)
  destructors, catch block features. Clearly we would have liked to give attention to
  enterprise logic and ignore the compiled library features, additionally combined
  into the binary

After these makes an attempt we determined to alter our method to slice the DLL based mostly
on purposeful space/workflow quite than take into account the whole meeting code.

Discovering the related operate

The primary problem within the purposeful space / workflow method is to discover a
hyperlink or entry level among the many 1000s of features.

One of many out there choices was to rigorously take a look at the constants and
strings within the DLL. We used the historic context: late Nineteen Nineties or early 2000
widespread architectural sample adopted in that interval was to insert knowledge into
the DB: was to both “choose for insert” or “insert/replace dealt with by saved
process” or through ADO (which is an ORM). Curiously we discovered all of the
patterns in several elements of the system.

Our performance was about inserting or updating the DB on the finish of the
course of however we could not discover any insert or replace queries within the strings, no
saved process to carry out the operation both. For the performance we
had been on the lookout for, it occurred to really use a SELECT by means of SQL after which
up to date through ADO (activex knowledge object microsoft library).

We acquired our break based mostly on the desk identify talked about within the
strings/constants, and this led to discovering the operate which is utilizing that
SQL assertion. Preliminary take a look at that operate did not reveal a lot, it may very well be
in the identical purposeful space however a part of a unique workflow.

Constructing the related subtree

ASM code, and our disassembly software, gave us the operate name reference
knowledge, utilizing it we walked up the tree, assuming the assertion execution is one
of the leaf features, we navigated to the mum or dad which referred to as this to
perceive its context. At every step we transformed ASM into pseudo code to
construct context.

Earlier after we transformed ASM to pseudocode utilizing brute-force we could not
cross confirm whether it is true. This time we’re higher ready as a result of we all know
to anticipate what may very well be the potential issues that would occur earlier than a
sql execution. And use the context that we gathered from earlier steps.

We mapped out related features utilizing this name tree navigation, generally
we now have to keep away from mistaken paths. We realized about context poisoning in a tough
approach, in-advertely we handed what we had been on the lookout for into LLM. From that
second LLM began colouring its output focused in the direction of what we had been trying
for, main into mistaken paths and eroding belief. We needed to recreate a clear
room for AI to work in throughout this stage.

We acquired a excessive stage define of what the totally different features had been, and what
they may very well be doing. For a given work move, we narrowed down from 4000+
features to 40+ features to take care of.

Multi-Go Enrichment

AI accelerated the meeting archaeology layer by layer, cross by cross: We
utilized multi cross enrichment. In every cross, we both navigated from leaf
node to prime of the tree or reverse, in every step we enriched the context of
the operate both utilizing its mother and father context or its little one context. This
helped us to alter the technical conversion of pseudocode right into a purposeful
specification. We adopted easy methods like asking LLM to offer
significant technique names based mostly on identified context. After a number of passes we construct
out your entire purposeful context.

Validating the entry level

The final and important problem was to substantiate the entry operate. Typical
to C++, digital features made it more durable to hyperlink entry features in school
definition. Whereas performance regarded full beginning with the foundation node,
we weren’t certain if there’s every other extra operation occurring in a
mum or dad operate or a wrapper. Life would have been simpler if we had debugger
enabled, a easy break level and assessment of the decision stack would have
confirmed it.

Nevertheless with triangulation methods, like:

1. Name stack evaluation
2. Validating argument signatures and the the return signature within the
   stack
3. Cross-checking with UI layer calls (e.g., associating technique signature
   with the “submit” name from Internet tier, checking parameter varieties and utilization, and
   validating in opposition to that context)