deslop
Regression-safe cleanup of AI-generated bloat in C# code. Deletion-first, requires green tests before starting.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
القائمة
Regression-safe cleanup of AI-generated bloat in C# code. Deletion-first, requires green tests before starting.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Root-cause native Bannerlord CTDs (AccessViolation in TaleWorlds.Native.dll) via Event Log offsets, offline disassembly, and live debugger forensics. No symbols needed.
Dispatch an independent Codex verification job directly via `codex exec` Bash call, then retrieve and present results
Inspect all changed files, group by logical concern, and guide atomic per-concern commits following TAOM's 50/72 rule and commit trailer convention.
Load a snapshot saved by /context-save and present it so this session can pick up where the last one left off without re-deriving decisions.
Launch parallel deep-dive agents to review completed work for quality, standards, compatibility, completeness, and cross-system data flow
Respond to a Bannerlord engine update (Steam force-bump or deliberate migration) — preserve the decompile baseline, regen, diff, re-verify bindings and creature data, control-battle.
| name | deslop |
| description | Regression-safe cleanup of AI-generated bloat in C# code. Deletion-first, requires green tests before starting. |
| argument-hint | ["path or feature to clean up","e.g. \"Main/Features/Execution\" or \"WargAttackService\""] |
Regression-safe cleanup of AI-generated bloat. Targets dead code, duplicate logic, over-abstraction, and unnecessary complexity — without changing behavior.
Scope: $ARGUMENTS (if empty, use git diff --name-only to find recently modified files)
dotnet test TAOM.Tests --no-build 2>&1 | tail -5
If tests are failing: STOP. Do not clean up broken code — you'll mask real bugs. Report the failures and exit.
Scan target files for these patterns. Do not change anything yet.
| Pattern | TAOM Example |
|---|---|
| Dead code | Private methods never called, unused fields, commented-out patch blocks |
| Redundant null guards | Null checks on DryIoc-injected services (the container guarantees non-null) |
| Restatement comments | // This method calculates the wage above CalculateWage() |
| Empty base calls | public Foo() : base() {} with no reason to exist |
| Adapter over-wrapping | Adapter that just delegates every property 1:1 with no translation logic |
| Single-use private methods | private bool IsValid() { return x != null; } called exactly once |
| Duplicate guard clauses | Same null/empty check copy-pasted across 3+ methods in the same class |
| Over-defensive catch blocks | try { } catch (Exception) { } around code that can't throw |
| Unused using directives | using TaleWorlds.Library; when nothing from that namespace is used |
| Harmony patch boilerplate noise | __state parameters declared but never used, __result assigned then immediately returned unchanged |
For each item found, record: file:line, pattern type, deletion safety (SAFE / RISKY).
Mark RISKY if:
Apply changes in this order — safest first:
using directives (compiler-verified safe)grep -r "MethodName" Main/ first)Hard limits — never do these:
#pragma warning suppressions without understanding why they existdotnet test TAOM.Tests --no-build 2>&1 | tail -10
If any test fails: revert the last change (git diff to identify it), report which deletion broke which test, and mark that item RISKY in the report. Do not delete it.
DESLOP REPORT
=============
Scope: [files or feature name]
Files touched: N
Lines deleted: N
Lines added: N (extractions only)
CLEANED:
- [file:line] [pattern] — [brief description]
- ...
SKIPPED (RISKY):
- [file:line] [pattern] — [reason skipped]
Tests: PASS (N tests) / FAIL (list failures)
NET: [clean / needs manual review of N risky items]