Back to skills
SKILL.md
Detecting Memory Corruption
ASecurityDetects various memory corruption vulnerabilities beyond simple buffer overflows including heap overflow, stack smashing, and double free. Use when analyzing complex memory management, heap operations, or investigating memory safety issues.
- 666 stars
- 0 votes
- 0 copies
- 5 views
- Added February 8, 2026
Security analysis
100/100npx -y skills add majiayu000/claude-skill-registry --skill detecting-memory-corruption --agent claude-codeAre you the author of Detecting Memory Corruption?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/majiayu000-detecting-memory-corruption)---
name: detecting-memory-corruption
description: Detects various memory corruption vulnerabilities beyond simple buffer overflows including heap overflow, stack smashing, and double free. Use when analyzing complex memory management, heap operations, or investigating memory safety issues.
---
# Memory Corruption Detection
## Detection Workflow
1. **Identify memory operations**: Find all malloc/free/realloc calls, stack buffer allocations, pointer arithmetic operations
2. **Trace memory usage**: Use `xrefs_to` to trace data, track pointer values through code, identify all access patterns
3. **Check bounds**: Verify size calculations, check array index validation, assess pointer arithmetic safety
4. **Assess impact**: Can corruption control execution? Is there information disclosure? Can corruption cause DoS?
## Key Patterns
- Heap overflow: malloc() followed by unchecked writes, off-by-one errors, integer overflow in size calculations
- Stack smashing: large stack buffer allocations, unchecked array access, variable-length arrays, return address overwrites
- Pointer arithmetic: pointer arithmetic without bounds checking, array indexing with user-controlled indices, out-of-bounds pointer access
- Memory leaks: missing free() after malloc(), reference cycles, exception paths skipping cleanup
## Output Format
Report with: id, type, subtype, severity, confidence, location, allocation (function, address, size), corruption (function, address, source), vulnerability, root cause, exploitable, attack scenario, mitigation.
## Severity Guidelines
- **CRITICAL**: Memory corruption allowing code execution
- **HIGH**: Memory corruption with data corruption or DoS
- **MEDIUM**: Memory leaks or minor corruption
- **LOW**: Potential issues with limited impact
## See Also
- `patterns.md` - Detailed detection patterns and exploitation scenarios
- `examples.md` - Example analysis cases and code samples
- `references.md` - CWE references and mitigation strategiesAttribution
Comments
Loading comments…
Newsletter
Get the weekly list of safe new skills, and the ones we pulled.