Detect silent fallback patterns and error suppression in code.
npx skills add https://github.com/d3ming/dot-agents --skill fallback-hunter
Please help me install this Agent Skill. Skill: fallback-hunter Source: https://github.com/d3ming/dot-agents/tree/main/master/skills/fallback-hunter Command: npx skills add https://github.com/d3ming/dot-agents --skill fallback-hunter
This skill hunts code that silently replaces missing or bad data with defaults or swallows errors, which creates plausible-but-wrong outcomes in data-heavy systems and financial calculations. It helps teams locate hidden error suppression and fail-soft anti-patterns so critical data paths fail fast and loud instead of corrupting decisions.
.fillna(0) or broad excepts that could silently change portfolio valuations or risk signals.Scan the src/ tree for silent fallbacks and generate a prioritized P0–P3 remediation report.
High-intent search queries and answers about installing and using this skill.
To detect silent fallbacks, scan your repository for patterns like `.fillna(0)`, default dict.get, and broad exception handlers that mask failures. This produces a file-and-line-level report prioritizing hidden error suppression in scoring, pricing, and valuation code.
A silent fallback pattern is code that replaces missing or bad data with defaults, creating plausible-but-wrong outcomes. It affects data quality by allowing fail-soft anti-patterns to corrupt decisions in data-heavy systems instead of failing fast and loud.
Audit your codebase by scanning the `src/` tree for falsy-default chains, missing-key mappings, and try/except fallbacks. The scan categorizes findings into P0–P3 severity tiers based on whether the fallback affects critical paths like portfolio valuation or only display code.
Yes, you can generate suggested semgrep rules for prevention. After identifying silent fallbacks and broad excepts, the analysis provides concrete prevention suggestions including semgrep rules and code-review gates to stop hidden error suppression in data-processing pipelines.
Yes, this approach works specifically for auditing risk-calculation services and trading systems. It scans financial repositories to find `.fillna(0)` or broad excepts that could silently change portfolio valuations or risk signals, classifying them by decision-path severity.
The best way to prioritize remediation is using context-aware classification tiers from P0 to P3. Findings affecting scoring, pricing, and portfolio valuation receive higher priority over display code, producing a prioritized report with code snippets and hidden-effect analysis.
DonTiziAutomate security analysis to detect secrets, vulnerabilities, and insecure patterns.
itallstartedwithaideaCombine CodeQL data-flow queries and Semgrep rules to detect vulnerabilities.
prof-danielcRun static analysis tools and aggregate results into a prioritized issue report.
massgenLocate text and structural code patterns across large codebases using ripgrep and ast-grep.