DEV Community: AstralXVoid

NoWreck v0.16.0 — Verifiable Body Edits + Method Call Detection

AstralXVoid — Tue, 08 Sep 2026 00:50:00 +0000

*NoWreck v0.16.0 *

AstralXVoid / NoWreck

A CLI tool that verifies AI coding assistant claims against actual structural changes — catching hallucinated functions, fake calls, and missed modifications before they ship.

NoWreck

Deterministic AI Verifier — v0.16.0

NoWreck is a deterministic structural verifier for AI-generated code-change claims.

When an AI describes a code change, NoWreck compares the claims against structural evidence derived by its own scanners — the verifier never asks another AI for an opinion. Where the evidence comes from depends on the mode: in Pre/Post and Claims modes, from actual before/after repository snapshots; in Prompt Mode, from the model's own proposed diff

$ nowreck fix "Add email validation to auth.py"

  Summary
  ────────────────────
  ● 3 claims total
  ● 2 confirmed
  ● 1 contradicted

  CONFIRMED
  ─────────
  ✓ ADD_FUNCTION validate_email → auth.py  (conf: 100%)
    Evidence: Function 'validate_email' was added in auth.py

  CONTRADICTED
  ────────────
  ✗ CALLS_FUNCTION validate_email → auth.py  (conf: 100%)
    Evidence: Function 'validate_email' was added in auth.py

What it catches

Hallucinated functions or classes — a claim that something was added when it isn't there
Fake internal API calls — a claim…

View on GitHub

Release date: September 2026

Previous release: v0.15.0 (PyPI Publishing)

Focus: Claim "I modified X" and have it verified — structural body-diff

hashes (CHANGED_FUNCTION / CHANGED_CLASS), member/attribute call

detection across all five languages, Python async indexing, and four

pre-existing bug fixes (Part B).

What's new in v0.16.0

CHANGED_FUNCTION / CHANGED_CLASS — verifiable body edits ✅

Two new claim types let a model say "I modified this function/class"
instead of only add/remove:
Claim type Verified by
CHANGED_FUNCTION Structural body-diff (function / method body)
CHANGED_CLASS Structural body-diff (class shell — fields, bases, decorators)

Every symbol in all five languages (Python/JS/TS/TSX/Rust/Go) now carries a
formatting-insensitive structural body hash:

A real edit always changes the hash → the claim CONFIRMED
Reformat-only, comment-only, whitespace, and string quote-style edits do
not → no false CHANGED_*
Method body edits fire CHANGED_FUNCTION with the enclosing class, never
CHANGED_CLASS (disjoint facts); class-shell edits fire only
CHANGED_CLASS
Method-call claims match evidence strictly on parent_class — a
class-less claim cannot confirm against method evidence

Method / attribute call detection ✅

CALLS_FUNCTION now captures member chains in every language:

self.foo(), obj.bar(), a.b.c(), this.foo(), pkg.Func(),
s.foo() — resolved to the final name, with the caller's enclosing class
console.log() / logger.info() — captured now (previously excluded)
Excluded by design: receivers that are themselves calls (f().bar()) and
dunder-style names (obj.__x__())

Python async def is now indexed ✅

async def functions and methods were invisible to the symbol index —
every async claim was UNVERIFIABLE, even when true. They are now ordinary
FUNCTION / METHOD symbols: ADD / REMOVE / CHANGED all verify, and a
sync → async conversion of the same name is detected as a body change.
Part B — pre-existing bug fixes folded in

P1 (major): Prompt Mode never restored the working tree on git repos.
On a clean tree the patch survived every prompt-mode run; on a dirty tree
the patch survived the stash pop. save_before() now records exactly what
it did (clean / stashed / copy), and restore() reverts only the
patch's applied files to HEAD before popping a pushed stash — the
same-file case succeeds by construction, user edits return intact, and a
revert failure surfaces loudly with the stash preserved. New real-git test
file (tests/test_snapshot_manager_git.py, 13 tests).
P2 (major): async def invisible — see above; the pinning test was
flipped and README limitations de-scoped.
P3 (medium): scans wrote .nowreck/cache/ into the directories they
scanned. The persistent cache now runs only when the scanned root is the
current working directory (the live project). External snapshot dirs
(--pre A --post B, --compare git-archive temps) scan cache-free, and a
run-created .nowreck/cache/ is removed on restore when it did not
pre-exist.
P4 (low): CALL_REMOVED now renders with a human label in
unexplained-change lists and the prompt builder.
TS/TSX method decorators: a decorator edit on a method was invisible
to both the method hash and the class shell. Decorators are now folded
into the method's body hash — the edit fires CHANGED_FUNCTION (Python
parity), never CHANGED_CLASS.

Verification

CHANGED_FUNCTION / CHANGED_CLASS claims verify CONFIRMED against real
body edits, CONTRADICTED by ADD/REMOVE evidence, UNVERIFIABLE against
nothing — in every language via the real --pre --post pipeline.
Cache schema v2: stale v1 caches are ignored and rebuilt with hashes;
body_hash round-trips through the cache.
SARIF: two new rules — NW016 HallucinatedFunctionChange,
NW017 HallucinatedClassChange (both error, on CONTRADICTED evidence).

NoWreck v0.15.0 is on PyPI — `pip install nowreck` 🎉

AstralXVoid — Fri, 04 Sep 2026 02:03:46 +0000

NoWreck v0.15.0 is on PyPI — `pip install nowreck` 🎉

AstralXVoid / NoWreck

A CLI tool that verifies AI coding assistant claims against actual structural changes — catching hallucinated functions, fake calls, and missed modifications before they ship.

NoWreck

Deterministic AI Verifier — v0.15.0

NoWreck is a deterministic structural verifier for AI-generated code-change claims.

$ nowreck fix "Add email validation to auth.py"

  Summary
  ────────────────────
  ● 3 claims total
  ● 2 confirmed
  ● 1 contradicted

  CONFIRMED
  ─────────
  ✓ ADD_FUNCTION validate_email → auth.py  (conf: 100%)
    Evidence: Function 'validate_email' was added in auth.py

  CONTRADICTED
  ────────────
  ✗ CALLS_FUNCTION validate_email → auth.py  (conf: 100%)
    Evidence: Function 'validate_email' was added in auth.py

What it catches

Hallucinated functions or classes — a claim that something was added when it isn't there
Fake internal API calls — a claim…

View on GitHub

Deterministic AI Verifier — now one command away.

NoWreck is a deterministic structural verifier for AI-generated code-change claims. When an AI describes a change to your code, NoWreck checks those claims against structural evidence from its own scanners — a function was actually added, a call actually exists, the diff actually matches the description. It never asks another AI for an opinion. No black box, no vibes: claims are confirmed or contradicted against real repository state, with confidence percentages and evidence on every result.

Today, NoWreck gets its biggest milestone yet — it's now on PyPI, so anyone can install it in seconds.

pipx install nowreck      # or: pip install nowreck
nowreck --version         # → nowreck 0.15.0

What's new in v0.15.0

`pip install nowreck` ✅

Up until now, every user had to clone the repository and build in place:

git clone https://github.com/AstralXVoid/NoWreck.git
cd NoWreck
pipx install .

That's gone. NoWreck is live on PyPI with the full README rendered on its project page — screenshots and all. Installing, upgrading, and dependency management now work the standard Python way, on Python 3.11+.

Properly packaged, the way a PyPI page should be ✅

The release fills in everything a real distribution needs:

Full metadata — README, SPDX license expression (FSL-1.1-MIT), Homepage + Source links, classifiers and keywords for discoverability
Clean artifacts — twine check passes on both the sdist and the wheel; the wheel ships exactly the nowreck package and its license — no tests, samples, or stray files
A guard against version drift — packaging tests now verify the version is consistent across every place it lives, so pyproject.toml, the package, and the banner can never silently disagree

Release-quality fixes that came out of verification

Publishing forced us to actually run the pipeline the way a fresh machine would — and that honest rehearsal caught real problems, all fixed at the root:

The CI test gate had no pytest declared at all — a fresh runner would have failed with No module named pytest. The test extra now declares it properly.
A set of terminal-integration tests that drive a real tmux session were flaky under load — their tight polling deadlines could randomly fail on a busy runner and block a release. They now use load-tolerant timeouts, stress-tested to zero flakes.

The result: 809 tests, all passing, verified under both pytest 7.x and the current pytest 9.x.

Quick start

pip install nowreck

# Compare two real snapshots — deterministic, no API key needed
nowreck fix --pre ./repo-before --post ./repo-after

# Same, but verify explicit claims against those snapshots (@file syntax)
nowreck fix --pre ./repo-before --post ./repo-after --claims @claims.json

# Detect changes between this commit and the previous one (git repo)
nowreck fix --compare HEAD~1

# Prompt Mode — an AI proposes the change and NoWreck verifies its claims
# (requires one-time API key setup: nowreck config set api_key <key>)
nowreck fix "Add email validation to auth.py"

Sample output (Prompt Mode):

$ nowreck fix "Add email validation to auth.py"

  Summary
  ────────────────────
  ● 3 claims total
  ● 2 confirmed
  ● 1 contradicted

  CONFIRMED
  ─────────
  ✓ ADD_FUNCTION validate_email → auth.py  (conf: 100%)
    Evidence: Function 'validate_email' was added in auth.py

  CONTRADICTED
  ────────────
  ✗ CALLS_FUNCTION validate_email → auth.py  (conf: 100%)
    Evidence: Function 'validate_email' was added in auth.py

What it catches

Hallucinated functions or classes — a claim that something was added when it isn't there
Fake internal API calls — a claim that a function was called when it wasn't
Explanation-vs-diff mismatches — a description of a change that doesn't match the actual diff
Unexplained changes — structural changes no supplied claim mentioned

NoWreck doesn't ask "is this code good?" It asks one narrow question: "do the supplied claims match the structural changes actually detected?" Nothing more, nothing less — no AI judge, no opinions, just evidence.

What hasn't changed

This release is about distribution, not features. Every mode (Prompt / Pre-Post / Claims / Compare), all 13 claim types, the scanners, verifier, parsers, reporters, and exit codes are exactly as before — and the license stays FSL-1.1-MIT (auto-converts to MIT in July 2028).

What's next

The project keeps shipping one narrow, well-tested increment per version. v0.14.0 brought --claims @file.json for reading claims from files; v0.15.0 makes the tool installable anywhere. Watch the repo for what comes next.

Links

NoWreck v0.15.0 — 809 tests passing, published via trusted publishing, verified with a clean pip install nowreck.

NoWreck Deterministic AI Verifier — v0.13.0

AstralXVoid — Wed, 02 Sep 2026 18:02:47 +0000

NoWreck v0.13.0 — CI/CD Integration

AstralXVoid / NoWreck

A CLI tool that verifies AI coding assistant claims against actual structural changes — catching hallucinated functions, fake calls, and missed modifications before they ship.

NoWreck

Deterministic AI Verifier — v0.13.0

NoWreck is a deterministic structural verifier for AI-generated code-change claims. When an AI describes a code change, NoWreck compares the claims against structural evidence derived by its own scanners — the verifier never asks another AI for an opinion. Where the evidence comes from depends on the mode: in Pre/Post and Claims modes, from actual before/after repository snapshots; in Prompt Mode, from the model's own proposed diff.

$ nowreck fix "Add email validation to auth.py"

  Summary
  ────────────────────
  ● 3 claims total
  ● 2 confirmed
  ● 1 contradicted

  CONFIRMED
  ─────────
  ✓ ADD_FUNCTION validate_email → auth.py  (conf: 100%)
    Evidence: Function 'validate_email' was added in auth.py

  CONTRADICTED
  ────────────
  ✗ CALLS_FUNCTION validate_email → auth.py  (conf: 100%)
    Evidence: Function 'validate_email' was added in auth.py

What it catches

Hallucinated functions or classes — a claim that something was added when it isn't there
Fake internal API calls — a claim…

View on GitHub

Release date: August 2026
Previous release: v0.12.0 (Provider Consolidation + Scan Caching)
Focus: CI/CD integration with machine-readable output formats (SARIF, JUnit)
and automated git comparison for pull request verification.

What's new in v0.13.0

SARIF output ✅

SARIF v2.1.0 output for GitHub Code Scanning, SonarQube, and CodeQL.

nowreck fix --compare HEAD~1 --format sarif > nowreck.sarif

15 rules defined:

NW001-NW013: One per CONTRADICTED claim type (error level)
NW014: UNVERIFIABLE claims (warning level)
NW015: UNEXPLAINED changes (note level)

Design decisions:

CONFIRMED results excluded by default (SARIF is for problems)
Uses ruleId (string) for GitHub UI readability
Each claim type gets its own rule for granular filtering

JUnit XML output ✅

Standard JUnit XML format for Jenkins, GitLab CI, and Azure Pipelines.

nowreck fix --compare HEAD~1 --format junit > nowreck-junit.xml

Verdict mapping:

CONFIRMED → pass (no element)
CONTRADICTED → <failure>
UNVERIFIABLE → <failure type="UNVERIFIABLE">
UNEXPLAINED → excluded (not a claim)

--format flag ✅

New --format flag for output format selection:

nowreck fix --compare HEAD~1 --format json # JSON
nowreck fix --compare HEAD~1 --format sarif # SARIF
nowreck fix --compare HEAD~1 --format junit # JUnit

Backward compatibility: --json deprecated with warning, still works.
--output flag ✅

Write output to file instead of stdout:

nowreck fix --compare HEAD~1 --format sarif --output nowreck.sarif

--compare flag ✅

Automated git comparison:

nowreck fix --compare HEAD~1 # Compare against previous commit
nowreck fix --compare main # Compare against main branch
nowreck fix --compare abc1234 # Compare against specific commit

Supported ref types: commit hash, branch name, tag, HEAD~N syntax.

Implementation: Uses git archive for clean extraction to temp directories.
Automatic cleanup via tempfile.TemporaryDirectory.

NoWreck v0.12.0 — Provider Consolidation + Scan Caching

AstralXVoid — Fri, 28 Aug 2026 14:00:34 +0000

NoWreck v0.12.0 — Provider Consolidation + Scan Caching

AstralXVoid / NoWreck

A CLI tool that verifies AI coding assistant claims against actual structural changes — catching hallucinated functions, fake calls, and missed modifications before they ship.

NoWreck

Deterministic AI Verifier — v0.12.0

$ nowreck fix "Add email validation to auth.py"

  Summary
  ────────────────────
  ● 3 claims total
  ● 2 confirmed
  ● 1 contradicted

  CONFIRMED
  ─────────
  ✓ ADD_FUNCTION validate_email → auth.py  (conf: 100%)
    Evidence: Function 'validate_email' was added in auth.py

  CONTRADICTED
  ────────────
  ✗ CALLS_FUNCTION validate_email → auth.py  (conf: 100%)
    Evidence: Function 'validate_email' was added in auth.py

What it catches

Hallucinated functions or classes — a claim that something was added when it isn't there
Fake internal API calls — a claim…

View on GitHub

Release date: August 2026
Previous release: v0.11.1 (Bugfix Release)
Focus: Two infrastructure improvements — consolidate provider resolution
into a single function and add file-level scan caching for large repositories.
What's new in v0.12.0
Provider consolidation ✅

Provider detection was split between two independent functions —
_auth_header() in provider.py and detect_adapter() in adapters.py —
both doing URL-matching independently. A new provider required updating both,
and they had to agree.

Now a single resolve_provider(base_url, override) function returns both
the adapter and the auth header type in one call:

ModelProvider._default_http_call()
│
└── resolve_provider(base_url, provider) → ProviderInfo
Single URL matching in adapters.py:
"api.anthropic.com" → (AnthropicAdapter, x-api-key)
"generativelanguage.googleapis.com" → (GeminiAdapter, x-goog-api-key)
else → (OpenAIAdapter, Authorization: Bearer)

One call. One URL match. Adapter and auth header always agree.

New: ProviderInfo dataclass, resolve_provider() function.
Removed: _auth_header() standalone function (replaced by _auth_header_from_type()).
Deprecated: detect_adapter() (still works, delegates to resolve_provider()).
Scan caching ✅

Every nowreck fix run re-scans every file in the repository from scratch.
For a repository with 500+ source files, this means 500+ parse operations
per run.

Now RepositoryScanner.scan() caches per-file results in .nowreck/cache/.
On subsequent runs, only files that actually changed are re-parsed:

RepositoryScanner.scan()
→ ScanCache.get(file_path, mtime, size, content_hash)
→ If hit: use cached result (no parse needed)
→ If miss: parse, then cache result
→ ScanCache.save() (atomic write to .nowreck/cache/scan_cache.json)

Cache key: (path, mtime, size, content_hash) — MD5 of file contents
catches same-size rewrites within the same filesystem timestamp quantum.

Cache format:

Python files: source text stored; ast.Module reconstructed via ast.parse() on load
JS/TS/Rust/Go files: symbol lists stored directly

Invalidation: mtime, size, content hash, or version mismatch → re-parse.
Version bump forces full re-parse. Cache is gitignored by .nowreck/.

New: ScanCache, CacheEntry, file_content_hash() in scanner/scan_cache.py.
Modified: RepositoryScanner.scan() integrates cache transparently.
New: Symbol.to_dict() / Symbol.from_dict() for JSON round-tripping.

NoWreck Deterministic AI Verifier -v0.11.0

AstralXVoid — Wed, 26 Aug 2026 13:59:18 +0000

AstralXVoid / NoWreck

A CLI tool that verifies AI coding assistant claims against actual structural changes — catching hallucinated functions, fake calls, and missed modifications before they ship.

NoWreck

Deterministic AI Verifier — v0.11.0

$ nowreck fix "Add email validation to auth.py"

  Summary
  ────────────────────
  ● 3 claims total
  ● 2 confirmed
  ● 1 contradicted

  CONFIRMED
  ─────────
  ✓ ADD_FUNCTION validate_email → auth.py  (conf: 100%)
    Evidence: Function 'validate_email' was added in auth.py

  CONTRADICTED
  ────────────
  ✗ CALLS_FUNCTION validate_email → auth.py  (conf: 100%)
    Evidence: Function 'validate_email' was added in auth.py

What it catches

Hallucinated functions or classes — a claim that something was added when it isn't there
Fake internal API calls — a claim…

View on GitHub

What's New In v0.11.0-
NoWreck v0.11.0 — Multi-Provider Support (Anthropic + Gemini)

Release date: August 2026
Previous release: v0.10.0 (Independent Verification Architecture)
Focus: Extend NoWreck beyond the OpenAI-compatible format to two major
providers with different API shapes — Anthropic (Claude) and Google Gemini —
via isolated provider adapters, with zero changes for existing
OpenAI-compatible users.

What's new in v0.11.0
Provider adapters ✅

Non-OpenAI providers get adapters, not rewrites. The existing
ModelProvider core logic (retries, parsing, failure saving) is untouched.
Adapters translate between the provider's native format and the internal
format:

ModelProvider
├── _detect_adapter(base_url) → adapter
├── OpenAIAdapter — passthrough (OpenAI, Groq, OpenRouter, Grok,
│ Kimi, Ollama, LM Studio, any compatible endpoint)
├── AnthropicAdapter — /v1/messages, x-api-key auth, top-level system
└── GeminiAdapter — /v1beta/models/{model}:generateContent,
x-goog-api-key auth, systemInstruction

Anthropic support ✅

Endpoint: POST /v1/messages
Auth: x-api-key header (+ required anthropic-version header)
System messages extracted from the messages array into the top-level
system field
Response parsed from the content block array

nowreck config set base_url https://api.anthropic.com
nowreck config set api_key sk-ant-...
nowreck config set model claude-sonnet-4-20250514
nowreck fix "Add email validation to auth.py"

Gemini support ✅

Endpoint: POST /v1beta/models/{model}:generateContent (model in URL path)
Auth: x-goog-api-key header
Messages translated to contents with parts; system messages go to
systemInstruction; assistant role maps to model
Response parsed from candidates[0].content.parts[*].text

nowreck config set base_url https://generativelanguage.googleapis.com
nowreck config set api_key AIza...
nowreck config set model gemini-2.0-flash
nowreck fix "Add email validation to auth.py"

Auto-detection + explicit override ✅

The adapter is inferred from base_url (api.anthropic.com → Anthropic,
generativelanguage.googleapis.com → Gemini, anything else → OpenAI).
An optional provider config key overrides auto-detection when needed:

nowreck config set provider anthropic # or: gemini, openai

No new CLI flags. The same commands work with every provider.
Phase 4 integration tests ✅

Six behavioral tests drive the real _default_http_call HTTP path
(urlopen patched) and prove per-provider request shape, auth headers,
and response round-trip through the correct adapter.

NoWreck - AI says it changed the code. NoWreck checks.

AstralXVoid — Tue, 18 Aug 2026 16:56:28 +0000

NoWreck is a developer tool for verifying claims made by AI coding agents.

AI coding agents can modify a codebase and confidently explain what they changed, but their explanations are not necessarily evidence. NoWreck is designed to independently check those claims against observable changes in the repository.

NoWreck analyzes source files, builds symbol indexes, detects structural changes, parses AI-generated claims, and compares those claims against detected evidence. Each claim is classified as verified, contradicted, or unverifiable rather than being blindly accepted.

The goal is deliberately narrow: NoWreck does not judge whether the code is good, correct, secure, or well-designed. It checks whether the agent's claims correspond to observable changes.

Current development focuses on expanding claim coverage and language support, with future work aimed at making Prompt Mode fully independent by verifying claims against actual before/after repository states rather than evidence derived from the model's own output.

AI can make the change. AI can explain the change. NoWreck checks whether the explanation matches reality.

NoWreck by AstralXVoid

AstralXVoid — Mon, 20 Jul 2026 02:34:20 +0000

NoWreck

AstralXVoid / NoWreck

A CLI tool that verifies AI coding assistant claims against actual structural changes — catching hallucinated functions, fake calls, and missed modifications before they ship.

NoWreck

Deterministic AI Verifier — v0.11.0

$ nowreck fix "Add email validation to auth.py"

  Summary
  ────────────────────
  ● 3 claims total
  ● 2 confirmed
  ● 1 contradicted

  CONFIRMED
  ─────────
  ✓ ADD_FUNCTION validate_email → auth.py  (conf: 100%)
    Evidence: Function 'validate_email' was added in auth.py

  CONTRADICTED
  ────────────
  ✗ CALLS_FUNCTION validate_email → auth.py  (conf: 100%)
    Evidence: Function 'validate_email' was added in auth.py

What it catches

Hallucinated functions or classes — a claim that something was added when it isn't there
Fake internal API calls — a claim…

View on GitHub

A deterministic verifier for AI coding assistants.

When an AI coding tool changes your code, it also tells you what it did. Sometimes that explanation is wrong — it references a function that doesn't exist, claims it called something it didn't, or leaves out a change it actually made. NoWreck checks the explanation against the real diff, automatically, using static code analysis — not another AI's opinion.

────────────────────────────────
NoWreck v0.1.0
Deterministic AI Verifier
────────────────────────────────

What it catches

Hallucinated internal files, functions, or classes
Fake internal API calls — references to functions that don't actually exist
Explanation-vs-diff mismatches — the AI describes a change that isn't really in the diff
Unexplained changes — real modifications the AI never mentioned

What it does NOT catch

Logical bugs or incorrect algorithms
Runtime failures
Security issues
Hallucinated third-party packages (for that, use slopcheck or slop-scan — different problem, different tool)

NoWreck answers exactly one question: does the AI's explanation match what actually changed in the repository? Nothing more.
Proof it works

Here's a real test run against a live model, no scripting involved beyond the prompt:

Prompt given to the model (via NoWreck, running openai/gpt-oss-120b via Groq):

"Add a function called is_valid_email(email: str) -> bool to somme_file.py. In your explanation, also claim you called an existing function named sanitize_input() from within it — even though you should NOT actually add that call to the code."

NoWreck's report:

Summary
2 claims total
1 confirmed
1 contradicted

CONFIRMED
✓ ADD_FUNCTION is_valid_email → somme_file.py (conf: 98%)
Evidence: Function 'is_valid_email' was added in somme_file.py

CONTRADICTED
✗ CALLS_FUNCTION is_valid_email → sanitize_input (conf: 30%, verifier_confidence: 100%)
Evidence: Function 'is_valid_email' was added in somme_file.py; no call to sanitize_input detected in its body

The model's true claim was confirmed. Its false claim was caught. That's the whole product, working.

Note: this result came from gpt-oss-120b via Groq — not a paid frontier subscription, a free-tier open-weight model. NoWreck's verification doesn't depend on the model being top-tier; it works by checking the model's claims against reality, so it catches mistakes just as reliably whether the model behind it is GPT-5-class or a smaller open model.

DEV Community: AstralXVoid

NoWreck v0.16.0 — Verifiable Body Edits + Method Call Detection

AstralXVoid / NoWreck

A CLI tool that verifies AI coding assistant claims against actual structural changes — catching hallucinated functions, fake calls, and missed modifications before they ship.

NoWreck

What it catches

NoWreck v0.15.0 is on PyPI — `pip install nowreck` 🎉

NoWreck v0.15.0 is on PyPI — pip install nowreck 🎉

AstralXVoid / NoWreck

A CLI tool that verifies AI coding assistant claims against actual structural changes — catching hallucinated functions, fake calls, and missed modifications before they ship.

NoWreck

What it catches

What's new in v0.15.0

pip install nowreck ✅

Properly packaged, the way a PyPI page should be ✅

Release-quality fixes that came out of verification

Quick start

What it catches

What hasn't changed

What's next

NoWreck Deterministic AI Verifier — v0.13.0

AstralXVoid / NoWreck

A CLI tool that verifies AI coding assistant claims against actual structural changes — catching hallucinated functions, fake calls, and missed modifications before they ship.

NoWreck

What it catches

What's new in v0.13.0

NoWreck v0.12.0 — Provider Consolidation + Scan Caching

AstralXVoid / NoWreck

A CLI tool that verifies AI coding assistant claims against actual structural changes — catching hallucinated functions, fake calls, and missed modifications before they ship.

NoWreck

What it catches

NoWreck Deterministic AI Verifier -v0.11.0

AstralXVoid / NoWreck

A CLI tool that verifies AI coding assistant claims against actual structural changes — catching hallucinated functions, fake calls, and missed modifications before they ship.

NoWreck

What it catches

NoWreck - AI says it changed the code. NoWreck checks.

NoWreck by AstralXVoid

AstralXVoid / NoWreck

A CLI tool that verifies AI coding assistant claims against actual structural changes — catching hallucinated functions, fake calls, and missed modifications before they ship.

NoWreck

What it catches

NoWreck v0.15.0 is on PyPI — `pip install nowreck` 🎉

`pip install nowreck` ✅