dotnet/skills/plugins/dotnet-diag/skills/android-tombstone-symbolication/SKILL.md
android-tombstone-symbolication
Symbolicate the .NET runtime frames in an Android tombstone file. Extracts BuildIds and PC offsets from the native backtrace, downloads debug symbols from the Microsoft symbol server, and runs llvm-symbolizer to produce function names with source file and line numbers. USE FOR triaging a .NET MAUI or Mono Android app crash from a tombstone, resolving native backtrace frames in libmonosgen-2.0.so or libcoreclr.so to .NET runtime source code, or investigating SIGABRT, SIGSEGV, or other native sign
- Source repository stars
- 4,922
- Declared platforms
- 0
- Static risk flags
- 2
- Last source update
- 2026-08-04
- Source checked
- 2026-08-04
Decision brief
What it does—and where it fits
Resolves native backtrace frames from .NET Android app crashes (MAUI, Xamarin, Mono) to function names, source files, and line numbers using ELF BuildIds and Microsoft's symbol server.
Not for
- Missing BuildIds: Logcat tombstones often omit BuildIds. Recover via: adb shell readelf -n /path/to/lib.so, CI build artifacts, or the runtime NuGet package (/.dotnet/packs/Microsoft.NETCore.App.Runtime.Mono.android-arm…
- Symbols not found (404): Pre-release/internal builds may not publish symbols. Check for local unstripped .so/.so.dbg in build artifacts or the NuGet runtime pack.
Compatibility matrix
Platform support, with evidence labels
| Platform | Status | Evidence | What to check |
|---|---|---|---|
| Codex | Not declared | No explicit evidence | Portability before use |
| Claude Code | Not declared | No explicit evidence | Portability before use |
| Cursor | Not declared | No explicit evidence | Portability before use |
| Gemini CLI | Not declared | No explicit evidence | Portability before use |
Installation
Inspect first. Install second.
The source command is displayed only when detected. A safe inspection prompt is always available so your agent can explain every action before execution.
npx skills add https://github.com/dotnet/skills --skill "plugins/dotnet-diag/skills/android-tombstone-symbolication"Inspect the Agent Skill "android-tombstone-symbolication" from https://github.com/dotnet/skills/blob/805a42a675a47f14fdd77a54aa474fcb8e499b9e/plugins/dotnet-diag/skills/android-tombstone-symbolication/SKILL.md at commit 805a42a675a47f14fdd77a54aa474fcb8e499b9e. List every install step, command, network request, credential, file read/write, external action, and rollback step. Explain whether it fits my task. Do not install or execute anything until I approve.
Workflow
What the source asks the agent to do
- 01
Workflow
Each backtrace frame has this format:
The script auto-detects NN pc frame lines with or without a backtrace: header, and strips logcat timestamp/tag prefixes automatically.Logcat-captured tombstones often omit BuildIds. Recover via adb shell readelf -n, CI build artifacts, or the .NET runtime NuGet package.GitHub issue pastes may mangle 1 pc into issue links — replace org/repoN pc with N pc before saving to a file. - 02
Step 1: Parse the Tombstone Backtrace
Each backtrace frame has this format:
The script auto-detects NN pc frame lines with or without a backtrace: header, and strips logcat timestamp/tag prefixes automatically.Logcat-captured tombstones often omit BuildIds. Recover via adb shell readelf -n, CI build artifacts, or the .NET runtime NuGet package.GitHub issue pastes may mangle 1 pc into issue links — replace org/repoN pc with N pc before saving to a file. - 03
Step 2: Identify .NET Runtime Libraries
Filter frames to .NET runtime libraries:
Filter frames to .NET runtime libraries:NativeAOT: No libcoreclr.so or libmonosgen-2.0.so — the runtime is statically linked into the app binary (e.g., libMyApp.so). The libSystem..so BCL libraries remain separate and can be symbolicated via the symbol server…Skip libc.so, libart.so, and other Android system libraries unless the user specifically asks. - 04
Step 3: Download Debug Symbols
For each unique .NET BuildId, download debug symbols:
For each unique .NET BuildId, download debug symbols:Verify with file libmonosgen-2.0.so.debug — should show ELF 64-bit ... with debuginfo, not stripped. If the download returns 404 or HTML, symbols are not published for that build. Do not add or subtract library base add… - 05
Step 4: Symbolicate Each Frame
The /w/1/s/ prefix is the CI workspace root — the meaningful path starts at src/runtime/, mapping to dotnet/dotnet VMR.
The /w/1/s/ prefix is the CI workspace root — the meaningful path starts at src/runtime/, mapping to dotnet/dotnet VMR.
Permission review
Static risk signals and limitations
Network access
The documentation includes network, browsing, or remote request actions.
https://msdl.microsoft.com/download/symbols/_.debug/elf-buildid-sym-<BUILDID>/_.debugNetwork access
The documentation includes network, browsing, or remote request actions.
curl -sL "https://msdl.microsoft.com/download/symbols/_.debug/elf-buildid-sym-1eb39fc72918c7c6c0c610b79eb3d3d47b2f81be/_.debug" \Writes files
The documentation asks the agent to create, modify, or delete local files.
Flags: `-CrashingThreadOnly` (limit to crashing thread), `-OutputFile path` (write to file), `-ParseOnly` (report libraries/BuildIds/URLs without downloading), `-SkipVersionLookup` (skip runtime version identification).Evidence record
Why each signal appears
| Signal | Value | Evidence type | Meaning |
|---|---|---|---|
| Quality score | 88/100 | Computed | Documentation, specificity, maintenance, and trust rules |
| Repository stars | 4,922 | Source | Repository attention, not individual Skill quality |
| Compatibility | 0 platforms | Source | Declared in the catalog source record |
| Usage guide | automated source guide | Editorial | Generated or reviewed according to the visible evidence level |
Pinned source
Provenance and original SKILL.md
- Repository
- dotnet/skills
- Skill path
- plugins/dotnet-diag/skills/android-tombstone-symbolication/SKILL.md
- Commit
- 805a42a675a47f14fdd77a54aa474fcb8e499b9e
- License
- MIT
- Collected
- 2026-08-04
- Default branch
- main
View the original SKILL.md
Android Tombstone .NET Symbolication
Resolves native backtrace frames from .NET Android app crashes (MAUI, Xamarin, Mono) to function names, source files, and line numbers using ELF BuildIds and Microsoft's symbol server.
Inputs: Tombstone file or logcat crash output, llvm-symbolizer (from Android NDK or any LLVM 14+ toolchain), internet access for symbol downloads.
Do not use when: The crash is a managed .NET exception (visible in logcat with a managed stack trace), the crashing library is not a .NET component (e.g., libart.so), or the tombstone is from iOS.
Workflow
Step 1: Parse the Tombstone Backtrace
Each backtrace frame has this format:
#NN pc OFFSET /path/to/library.so (optional_symbol+0xNN) (BuildId: HEXSTRING)
Extract: frame number, PC offset (hex, already library-relative), library name, and BuildId (32–40 hex chars).
Symbolicate all threads by default (background threads like GC/finalizer often have useful .NET frames). The crashing thread's backtrace is listed first; additional threads appear after --- --- --- markers.
Format notes:
- The script auto-detects
#NN pcframe lines with or without abacktrace:header, and strips logcat timestamp/tag prefixes automatically. - Logcat-captured tombstones often omit BuildIds. Recover via
adb shell readelf -n, CI build artifacts, or the .NET runtime NuGet package. - GitHub issue pastes may mangle
#1 pcinto issue links — replaceorg/repo#N pcwith#N pcbefore saving to a file. - If the script fails to parse a format, fall back to manual extraction of
#NN pc OFFSET library.so (BuildId: HEX)tuples.
Step 2: Identify .NET Runtime Libraries
Filter frames to .NET runtime libraries:
| Library | Runtime |
|---|---|
libmonosgen-2.0.so | Mono (MAUI, Xamarin, interpreter) |
libcoreclr.so | CoreCLR (JIT mode) |
libSystem.*.so | .NET BCL native components (Native, Globalization.Native, IO.Compression.Native, Security.Cryptography.Native.OpenSsl, Net.Security.Native) |
NativeAOT: No libcoreclr.so or libmonosgen-2.0.so — the runtime is statically linked into the app binary (e.g., libMyApp.so). The libSystem.*.so BCL libraries remain separate and can be symbolicated via the symbol server. For the app binary itself, you need the app's own debug symbols.
Skip libc.so, libart.so, and other Android system libraries unless the user specifically asks.
Step 3: Download Debug Symbols
For each unique .NET BuildId, download debug symbols:
https://msdl.microsoft.com/download/symbols/_.debug/elf-buildid-sym-<BUILDID>/_.debug
curl -sL "https://msdl.microsoft.com/download/symbols/_.debug/elf-buildid-sym-1eb39fc72918c7c6c0c610b79eb3d3d47b2f81be/_.debug" \
-o libmonosgen-2.0.so.debug
Verify with file libmonosgen-2.0.so.debug — should show ELF 64-bit ... with debug_info, not stripped. If the download returns 404 or HTML, symbols are not published for that build. Do not add or subtract library base addresses — offsets in tombstones are already library-relative.
Step 4: Symbolicate Each Frame
llvm-symbolizer --obj=libmonosgen-2.0.so.debug -f -C 0x222098
Output:
ves_icall_System_Environment_FailFast
/__w/1/s/src/runtime/src/mono/mono/metadata/icall.c:6244
The /__w/1/s/ prefix is the CI workspace root — the meaningful path starts at src/runtime/, mapping to dotnet/dotnet VMR.
Step 5: Present the Symbolicated Backtrace
Combine original frame numbers with resolved function names and source locations:
#00 libc.so abort+164
#01 libmonosgen-2.0.so ves_icall_System_Environment_FailFast (mono/metadata/icall.c:6244)
#02 libmonosgen-2.0.so do_icall (mono/mini/interp.c:2457)
#03 libmonosgen-2.0.so mono_interp_exec_method (mono/mini/interp.c)
For unresolved frames (??), keep the original line with BuildId and PC offset.
Automation Script
scripts/Symbolicate-Tombstone.ps1 automates the full workflow:
pwsh scripts/Symbolicate-Tombstone.ps1 -TombstoneFile tombstone_01.txt -LlvmSymbolizer llvm-symbolizer
Flags: -CrashingThreadOnly (limit to crashing thread), -OutputFile path (write to file), -ParseOnly (report libraries/BuildIds/URLs without downloading), -SkipVersionLookup (skip runtime version identification).
Finding llvm-symbolizer
Check the Android NDK first: $ANDROID_NDK_ROOT/toolchains/llvm/prebuilt/*/bin/llvm-symbolizer or $ANDROID_HOME/ndk/*/toolchains/llvm/prebuilt/*/bin/llvm-symbolizer. Also available via brew install llvm, apt install llvm, or xcrun --find llvm-symbolizer on macOS.
If unavailable, complete steps 1–3 and present the download commands and llvm-symbolizer commands for the user to run. Do not spend time installing LLVM.
Understanding the Output
CI source paths use these prefixes:
| Path prefix | Maps to |
|---|---|
/__w/1/s/src/runtime/ | src/runtime/ in dotnet/dotnet VMR |
/__w/1/s/src/mono/ | src/mono/ in the VMR (older builds) |
/__w/1/s/ | VMR root |
Runtime Version Identification
The script identifies the exact .NET runtime version by matching BuildIds against locally-installed runtime packs. It searches: SDK packs ($DOTNET_ROOT/packs/), NuGet cache (~/.nuget/packages/), and NuGet.org as an online fallback. When found, it extracts the version and source commit from the .nuspec <repository commit="..." /> element. Pass -SkipVersionLookup to disable. Requires llvm-readelf (auto-discovered from the NDK).
Validation
file <debug-file>showsELF ... with debug_info, not stripped- At least one .NET frame resolves to a function name (not
??) - Resolved paths contain recognizable .NET runtime structure (e.g.,
mono/metadata/,mono/mini/)
Stop Signals
- No .NET frames found: Report parsed frames and stop.
- All frames resolved: Present symbolicated backtrace. Do not trace into source or attempt to build/debug the runtime.
- Symbols not available (404): One attempt per BuildId, then stop. Report unsymbolicated frames with BuildIds and offsets.
- llvm-symbolizer not available: Use
-ParseOnly, present manual commands. Do not install LLVM.
Common Pitfalls
- Missing BuildIds: Logcat tombstones often omit BuildIds. Recover via:
adb shell readelf -n /path/to/lib.so, CI build artifacts, or the runtime NuGet package (~/.dotnet/packs/Microsoft.NETCore.App.Runtime.Mono.android-arm64/<version>/). Prefer pulling raw tombstone files (adb shell cat /data/tombstones/tombstone_XX) which always include BuildIds. - Symbols not found (404): Pre-release/internal builds may not publish symbols. Check for local unstripped
.so/.so.dbgin build artifacts or the NuGet runtime pack. - NativeAOT: No runtime
.soin the tombstone — runtime is in the app binary.libSystem.*.soBCL libraries still work with the symbol server; the app binary needs its own debug symbols. - Wrong llvm-symbolizer version: Use LLVM 14+ for best DWARF compatibility.
- Multiple BuildIds: Each .NET library has its own BuildId — download symbols for each separately.
Alternatives
Compare before choosing
alirezarezvani/claude-skills
app-store-optimization
App Store Optimization (ASO) toolkit for researching keywords, analyzing competitor rankings, generating metadata suggestions, and improving app visibility on Apple App Store and Google Play Store. Use when the user asks about ASO, app store rankings, app metadata, app titles and descriptions, app store listings, app visibility, or mobile app marketing on iOS or Android. Supports keyword research and scoring, competitor keyword analysis, metadata optimization, A/B test planning, launch checklist
dotnet/skills
migrate-vstest-to-mtp
Migrates .NET test projects from VSTest to Microsoft.Testing.Platform (MTP). Use when user asks to "migrate to MTP", "switch from VSTest", "enable Microsoft.Testing.Platform", "use MTP runner", set OutputType=Exe only for test projects in Directory.Build.props, or mentions EnableMSTestRunner, EnableNUnitRunner, or UseMicrosoftTestingPlatformRunner. USE FOR: MTP behavioral differences vs VSTest (exit code 8, zero tests discovered, --ignore-exit-code, TESTINGPLATFORM_EXITCODE_IGNORE); centralizing
HKUDS/Vibe-Trading
strategy-generate
Create, modify, and optimize quantitative trading strategies, then backtest and evaluate them.
K-Dense-AI/scientific-agent-skills
dask
Distributed computing for larger-than-RAM pandas/NumPy workflows. Use when you need to scale existing pandas/NumPy code beyond memory or across clusters. Best for parallel file processing, distributed ML, integration with existing pandas code. For out-of-core analytics on single machine use vaex; for in-memory speed use polars.