ReliabilityIQ

ReliabilityIQ is a repository risk-scanning toolkit that identifies portability and maintainability issues before cross-cloud deployment.

It combines static analysis, repository analytics, and incident correlation to surface high-signal findings in a single scan pipeline.

Purpose

ReliabilityIQ targets codebases that move across cloud platforms (initial focus: Microsoft EV2 and Azure DevOps ecosystems), across languages and artifacts, and highlights blockers that make cloud portability or reliable operations harder.

Primary focus areas:

• Cross-cloud portability blockers (hardcoded endpoints, IPs, paths, region/resource identifiers, etc.).
• Magic-string opportunity detection (high-signal ranking, not just raw pattern matching).
• Git churn vs. staleness hotspots with noise suppression for generated/readme/IDE artifacts.
• Incident-to-module/service association for operational risk concentration.
• Configurable/rule-driven scanning with optional plugin extension points.
• Web/table + heatmap visualization from persisted scan results.

Non-Goals (initial)

• Auto-fixing code.
• Full behavioral/runtime analysis (profiling, tracing).
• Zero-false-positive detection.
• Universal build-system and platform coverage in the first release.

High-Level Design

ReliabilityIQ is planned as a multi-component .NET solution:

• ReliabilityIQ.Cli: orchestrates scans, runs analyzers, writes results, and emits report summaries.
• ReliabilityIQ.Core: rule engine, finding model, classification, scoring, persistence abstraction.
• ReliabilityIQ.Analyzers.*: language + artifact + history/anomaly analyzers.
• ReliabilityIQ.Web: ASP.NET web UI over the same SQLite result store.

The scan pipeline:

1. Load config
2. Discover repository and classify files (source, config, docs, generated, vendor, IDE, etc.)
3. Run analyzers
4. Normalize + score findings with suppressions and allowlists
5. Persist findings/results into SQLite
6. Emit CLI summary and support web/dashboard consumption

Planned Technology Choices

• CLI: System.CommandLine
• Source parsing: Roslyn (C#), Tree-sitter via P/Invoke (C++/Python/PowerShell/Rust), plus JSON/YAML parsers where needed
• Git mining: LibGit2Sharp
• Persistence: Microsoft.Data.Sqlite + Dapper
• Web UI stack: ASP.NET, D3.js, Chart.js, DataTables (CDN-first bootstrap approach)
• Rule/config format: YAML with schema validation

Repository scanning output

• Filterable findings list
• Repo/module/file heatmap for hotspot analysis
• Configurable severity/priority and ranking
• Incident and changelog context for each relevant finding

Rule & configuration model

Planned rule/config structure:

• .reliabilityiq/config.yaml
• .reliabilityiq/rules/{portability, magic-strings, churn, incidents, deploy-ev2, deploy-ado}.yaml
• .reliabilityiq/rules/custom/*.yaml
• .reliabilityiq/allowlists/*.yaml

Precedence model:

1. Built-in defaults
2. Repo rules
3. Custom overrides
4. CLI flags (highest priority)

Also planned:

• Rule validation command (reliabilityiq rules validate)
• Inline and file-level suppressions for noise reduction
• Strong defaults for generated/vendor/IDE ignores with opt-in overrides

Current state note

This repository’s README is now aligned to the design intent in Docs/designdoc.md; implementation status may vary by branch and milestone.

CLI Reference (Current Implementation)

The CLI is a System.CommandLine app under ReliabilityIQ.Cli.

Usage

• dotnet run --project ReliabilityIQ.Cli -- <command> [options]
• If you have a built binary, use the executable in place of dotnet run.

Commands and arguments

scan portability

• --repo (required): repository path to scan
• --db: SQLite file path (default: <repo-root>/reliabilityiq-results.db)
• --suppressions: suppression file path (default: <repo-root>/reliabilityiq.suppressions.yaml)
• --fail-on: fail command with exit code 1 when findings at or above severity are present (error|warning|info)

scan magic-strings

• --repo (required): repository path to scan
• --db: SQLite file path (default: <repo-root>/reliabilityiq-results.db)
• --min-occurrences: minimum candidate occurrence count
• --top: max number of ranked candidates to persist
• --config: optional magic strings config (default: <repo-root>/reliabilityiq.magicstrings.yaml)

scan churn

• --repo (required): repository path to scan
• --db: SQLite file path (default: <repo-root>/reliabilityiq-results.db)
• --since: Git lookback window (for example 90d, 180d, 365d)
• --service-map: service boundary mapping file (format: ServiceName=glob)

scan deploy

• --repo (required): repository path to scan
• --db: SQLite file path (default: <repo-root>/reliabilityiq-results.db)
• --ev2-path-markers: semicolon-delimited EV2 markers override
• --ado-path-markers: semicolon-delimited ADO markers override

scan config-drift

• --repo (required): repository path to scan
• --db: SQLite file path (default: <repo-root>/reliabilityiq-results.db)

scan deps

• --repo (required): repository path to scan
• --db: SQLite file path (default: <repo-root>/reliabilityiq-results.db)

scan hygiene

• --repo (required): repository path to scan
• --db: SQLite file path (default: <repo-root>/reliabilityiq-results.db)

scan all

• --repo (required): repository path to scan
• --db: SQLite file path (default: <repo-root>/reliabilityiq-results.db)
• --fail-on: fail on error|warning|info for portability findings
• --suppressions: suppression file path (default: <repo-root>/reliabilityiq.suppressions.yaml)
• --min-occurrences: minimum magic-string candidate occurrences
• --top: max ranked magic-string candidates to persist
• --config: optional magic strings config (default: <repo-root>/reliabilityiq.magicstrings.yaml)
• --since: churn lookback window (for example 90d, 180d, 365d)
• --service-map: service boundary mapping file (format: ServiceName=glob)
• --ev2-path-markers: semicolon-delimited EV2 markers override
• --ado-path-markers: semicolon-delimited ADO markers override

rules validate

• --config: optional path to repo root or .reliabilityiq directory

rules list

• --config: optional path to repo root or .reliabilityiq directory
• --enabled-only: show only enabled rules
• --category: filter by category (portability, magic-strings, churn, deploy-ev2, deploy-ado, config-drift, dependencies, incidents, custom)

rules init

• --repo: initialize rule/config templates in repo root (default: current directory)

init

• --repo: initialize ReliabilityIQ configuration/templates (default: current directory)

server start

• --db (required): SQLite database file path
• --port: Kestrel HTTP port (default: 5100)
• --no-browser: do not auto-open browser

Example commands

# run portability scan on current repo
dotnet run --project ReliabilityIQ.Cli -- scan portability --repo . --db ./reliabilityiq-results.db

# run all scans with custom churn lookback and fail-on threshold
dotnet run --project ReliabilityIQ.Cli -- scan all \
  --repo . \
  --since 180d \
  --fail-on warning \
  --top 250 \
  --db ./reliabilityiq-results.db

# run only magic strings scan and persist top 100 candidates
dotnet run --project ReliabilityIQ.Cli -- scan magic-strings --repo . --top 100 --min-occurrences 3

# validate config in a repo
dotnet run --project ReliabilityIQ.Cli -- rules validate --config .

# start web server against a scan database
dotnet run --project ReliabilityIQ.Cli -- server start --db ./reliabilityiq-results.db --port 5200

# initialize .reliabilityiq templates
dotnet run --project ReliabilityIQ.Cli -- init --repo .