AtherosAI Compliance Kit Türkçe

Modules

Four modules over one core. Use one or all four. core imports nothing from them, and they import nothing from each other except a shared finding vocabulary — which is what lets you install one module's extras and not the rest.

M1 · atheros_kit.rag — corpus quality and bias

Measures the corpus, not the model. That is the gap: teams evaluate answers and never look at the knowledge base those answers are drawn from, and in a RAG system bias enters through the corpus far more often than through the weights.

CapabilityOutput
Connectorsin-memory, Chroma, pgvector, Pinecone, Milvus — read-only by construction
Chunk qualityempty, near-empty, exact and near duplicates, orphans, ragged dimensions, size outliers, PII
Semantic driftcentroid shift, per-dimension PSI, volume change — with a sample-size noise floor and an isotropy check
Biasseven dimensions, representation (four-fifths rule) and contextual framing → Fairness Score
Remediationordered recipes, emitted and never executed

Where it stops: lexicons are English-first and structural. assessed_dimensions and unassessable are separate fields, and a corpus nothing could be measured on scores None, not 100.

M2 · atheros_kit.guard — third-party API guardrails

The isolation layer between your application and an external model. Provider-agnostic: it wraps a callable, because every provider has a different client object and they all change, but str -> str does not.

CapabilityDetail
PII and custom entitiesStable placeholders (⟦EMAIL_1⟧) so the model can still tell two people apart. Reversible in-memory vault, or reversible=False for unrecoverable-by-construction.
ChecksumsLuhn, IBAN mod-97, TCKN, BSN — so an order number is not logged as a national identity number
Injection firewallInstruction override, prompt exfiltration, jailbreak personas, verdict manipulation, delimiter spoofing, tool smuggling, and encoded payloads that are decoded and re-scanned
Token governanceBudgets enforced *before* the call — a budget checked afterwards is a report, not a control
FallbackRetry → secondary → static, every step recorded, never silent

Three presets: observe() measures and blocks nothing; standard() masks and blocks the critical signatures; strict() drops the vault entirely and raises on anything suspicious.

Where it stops: the firewall is signature-based on purpose — a classifier that can be prompted is not a control. An LLM judge is available as a second opinion and never as the only one. Streaming interception is not supported in v1.

M3 · atheros_kit.euact — classification and Annex IV

CapabilityDetail
ClassificationStatutory order: Art. 5 → Annex I / Annex III → Art. 50 → minimal, against a versioned vocabulary
Grey zoneMultiple Annex III matches, sector conflicts, or GPAI plus high-risk lower confidence and name the conflict
ObligationsEach duty carries the module that produces its evidence — the spine that makes four modules one product
DossierAnnex IV sections 1–9, each covered / partial / missing, gaps printed as gaps
TransparencyArt. 50 zero-width marking, verification, C2PA-shaped metadata

Where it stops: it is a rule engine reading a description you wrote. It evaluates none of Art. 5's narrow exemptions and does not determine Art. 2 territorial scope.

M4 · atheros_kit.vendor — third-party and vendor risk

CapabilityDetail
24 criteriaCertifications, data protection, retention and training, security, AI transparency, operations — weighted
unknownPenalised at 20% credit, not skipped. A vendor cannot score well by declining to answer.
ResidencyAdequacy, SCCs, BCRs; the EU–US framework requires the *recipient's* certification, checked explicitly
Opt-outAvailable ≠ enabled ≠ contractual — three facts, reported separately
RegistrySeeded provider facts, dated, stale after 180 days and flagged as such by the tool itself

Where it stops: the seed registry is public documentation, which describes what a vendor offers — not what is configured on your account or written into your contract. Pass your own answers via overrides=.

atheros_kit.cicd — the gate

Compares what you measured to configured thresholds, writes atheros-report.json and a Markdown summary, and sets the exit code. Templates ship for GitHub Actions and GitLab CI.

Two rules: an unmeasured check never passes, and every skipped check is printed. Silent partial coverage reads as full coverage.

atheros_kit.iso — the ISO/IEC 42001 evidence pack

atheros-kit iso export --out ./evidence

Collects the hash-chained ledger into one document an auditor can read: the records grouped under the clause each one evidences, the chain verification, the clauses that hold no records, and the full list of what the Kit does not cover. It generates nothing — every row traces back to a ledger line by its digest.

Three properties make it usable as evidence rather than as marketing. It is a pure function of the ledger: no generation timestamp, no run id, so an auditor who doubts the document re-runs the command and diffs it. A broken chain is reported above the evidence, not beneath it, and the command exits 1 — there is no flag to silence that, because a flag to suppress the warning becomes the way the warning is suppressed. And an empty ledger is not a successful export: a command that exits green while producing a pack with no evidence has failed the operator at the moment it mattered.

Clause 6.1.2 is labelled *input only* wherever it appears. An EU AI Act classification is a regulatory categorisation; the clause asks for your own risk criteria, analysis and evaluation. See what this tool does not tell you.

atheros_kit.core — the foundations

Stdlib-only. Hash-chained audit ledger, layered config, the model-tier abstraction, the shared finding vocabulary, and report rendering in English and Turkish with redaction enforced at serialisation rather than left to callers.

· AtherosAI B.V.