How do I install the Opp Repl Sanitizer Tests skill?

Run npx skillmds@latest add tabgab/opp-repl-sanitizer-tests in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Opp Repl Sanitizer Tests safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Opp Repl Sanitizer Tests?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Opp Repl Sanitizer Tests free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Opp Repl Sanitizer Tests?

tabgab (@tabgab) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Opp Repl Sanitizer Tests by tabgab — Agent Skill

Name: Opp Repl Sanitizer Tests
Author: tabgab

Opp Repl Sanitizer Tests

Run simulations built with AddressSanitizer / UBSan instrumentation to catch memory errors and undefined behavior. Uses the `sanitize` build mode. Load when hunting down flaky crashes, UAFs, leaks, or UB that normal release builds miss.

tabgab a6b1a01 2.1 KB Updated Sep 17, 2026

File contents

Sanitizer tests

Sanitizer tests build the simulation with AddressSanitizer and UndefinedBehaviorSanitizer (sanitize build mode) and run it. Any detected issue is captured in stderr and reported as an ERROR.

Upstream reference: https://github.com/omnetpp/opp_repl/blob/main/doc/sanitizer_tests.md

Python API

run_sanitizer_tests(simulation_project=inet_project)

# Scope + longer CPU budget (sanitizer builds are slow)
run_sanitizer_tests(simulation_project=inet_project,
                    working_directory_filter="examples/ethernet",
                    cpu_time_limit="10s")

Command line

opp_run_sanitizer_tests --load inet.opp -p inet

Cost model

Sanitizer binaries run 2x-10x slower and use more memory. Keep cpu_time_limit tight and scope to affected areas. Nightly CI is a better fit than per-commit runs.

Interpreting failures

When ASAN reports, the error text appears in the result's stderr. print_stderr() on the failing TaskResult dumps it. Typical patterns:

heap-use-after-free / heap-buffer-overflow -- memory lifetime bug in a C++ module.
undefined-behavior from UBSan -- signed overflow, misaligned access, invalid enum cast, etc.
runtime error: load of null pointer -- dereferencing a NULL pointer during cleanup.

Pitfalls

Compiler and sanitizer-runtime versions affect output format; keep the toolchain stable across baseline and CI.
Some third-party libraries leak on purpose; suppress via LSAN_OPTIONS=suppressions=... when running.
Sanitizer builds cannot be used for speed tests or coverage — each is a separate build mode.

---name: opp-repl-sanitizer-testsdescription: Run simulations built with AddressSanitizer / UBSan instrumentation to catch memory errors and undefined behavior. Uses the `sanitize` build mode. Load when hunting down flaky crashes, UAFs, leaks, or UB that normal release builds miss.---# Sanitizer testsSanitizer tests build the simulation with AddressSanitizer andUndefinedBehaviorSanitizer (`sanitize` build mode) and run it. Anydetected issue is captured in stderr and reported as an ERROR.Upstream reference:https://github.com/omnetpp/opp_repl/blob/main/doc/sanitizer_tests.md## Python API run_sanitizer_tests(simulation_project=inet_project) # Scope + longer CPU budget (sanitizer builds are slow) run_sanitizer_tests(simulation_project=inet_project, working_directory_filter="examples/ethernet", cpu_time_limit="10s")## Command line opp_run_sanitizer_tests --load inet.opp -p inet## Cost modelSanitizer binaries run 2x-10x slower and use more memory. Keep`cpu_time_limit` tight and scope to affected areas. Nightly CI isa better fit than per-commit runs.## Interpreting failuresWhen ASAN reports, the error text appears in the result's`stderr`. `print_stderr()` on the failing `TaskResult` dumps it.Typical patterns:- `heap-use-after-free` / `heap-buffer-overflow` -- memory lifetime bug in a C++ module.- `undefined-behavior` from UBSan -- signed overflow, misaligned access, invalid enum cast, etc.- `runtime error: load of null pointer` -- dereferencing a NULL pointer during cleanup.## Pitfalls- Compiler and sanitizer-runtime versions affect output format; keep the toolchain stable across baseline and CI.- Some third-party libraries leak on purpose; suppress via `LSAN_OPTIONS=suppressions=...` when running.- Sanitizer builds cannot be used for speed tests or coverage — each is a separate build mode.## See also- `opp-repl-running-simulations` — underlying build/run machinery.- `opp-repl-tasks-and-results` — inspect stderr on ERROR.- `opp-repl-feature-and-release-tests` — run as part of a full suite.

Frequently asked questions