- Home
- Skills
- Code Quality & Review
- feature debt hunter
Works with the AI Agents you already use
feature debt hunter
Finds abandoned code, orphaned assets, and implementation gaps between docs and source code.
$5.99
feature debt hunter
Example session with this skill installed
Audit the current feature debt. We have a milestone next week and I need to know what's actually broken or half-finished in the IronBreach source.
- Read your context and instructions
- Compiled the feature debt hunter
- Generated the document
I have audited the Source/IronBreach/ and Content/ directories against the project brief. The report has been saved to Docs/FEATURE_DEBT_2023-10-27.md.
Summary
Found 14 debt items. The primary risks are the disconnected Sync overdrive finishers and three orphaned UI widgets in Content/UI/Disposable/.
- Finish: 4 items
- Cut: 6 items
- Keep: 2 items
- Ask: 2 items
Ranked items
| Feature/Asset | Status | File:Line / Path | Impact | Decision |
|---|---|---|---|---|
SyncOverdrive | Partial | Source/IB/Ability/Sync.cpp:142 | High | Finish: Add the delegate broadcast to the finisher loop. |
T_Mech_Old_Mask | Orphan | Content/Textures/Old/T_Mech_Old_Mask.uasset | Low | Cut: Delete; nothing in code or blueprints references this. |
IB.DebugGold | Unread | Source/IB/Player/IBCheat.cpp:12 | Med | Cut: CVar exists but AddGold logic was refactored. |
RaidPhase5 | Stub | Docs/Design/Raids.md | High | Ask: Connor - logic is missing from RaidManager, is this deferred? |
Promise gaps
- Striker Mech Class: Done. Full implementation found in
IBStriker.cpp. - Catastrophe Raids: Partial. Phases 1-4 functional; Phase 5 is currently a text stub in docs.
- Per-Player Loot: Absent. No implementation found in
IBLootTally.cpp.
Next steps
- Review the
Askitems with Connor regarding the Raid Phase 5 implementation. - Execute the
Cutlist for orphaned textures to reduce build size. - Hook the
SyncOverdrivedelegate to close the implementation gap before the milestone.
feature-debt-hunter.pdf
PDF · document
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Your codebase is full of features that are half-implemented, abandoned, or disconnected. This skill hunts down the gap between what your design docs promise and what your code actually executes, providing a prioritized cleanup roadmap.
What it does
- Identifies dead code by finding uncalled functions, unused enums, and console variables that no system reads.
- Audits asset orphans by checking for textures, materials, and maps that have no referencers in your content tree.
- Tracks promise gaps by comparing features listed in your design docs against their actual implementation status in the source.
- Detects stale work by surfacing untracked files, old git stashes, and folders marked disposable that were never deleted.
- Flags broken checks by identifying failing dev tests or checks that haven't run since their target modules were modified.
How it works
- Gather views: The agent scans design docs, handoff files, source code, and the content tree to build a project map.
- Probe: It runs concrete checks for markers like
TODO,deprecated, orunimplementedalongside deep greps for event bindings. - Verify: Every candidate for removal is cross-referenced against Blueprint bindings and data assets to prevent false positives.
- Rank: Each item is scored based on player-facing impact and implementation risk.
- Report: A structured debt report is generated with specific "Finish", "Cut", "Keep", or "Ask" decisions for every item.
Frameworks & tools
This skill works with C++ and Unreal Engine environments, utilizing git for tree state analysis and .uasset byte-searching for Blueprint bindings.
Why this beats prompting it yourself
Generic prompts often miss the subtle link between a C++ delegate and a Blueprint binding, leading to dangerous "dead code" suggestions. This skill specifically looks for cross-domain orphans that standard linters and LLM windows ignore.
Use cases
- Pre-milestone audit to ensure no stubbed features or placeholders make it into a demo.
- Project handoff to give a new developer a clear list of what is actually finished vs. what is just a doc promise.
- Refactoring phase to safely prune abandoned systems and unused assets.
- Cleanup after a long sprint where temporary hacks and
TODOmarkers have accumulated.
Known limitations
This skill does not delete files or commit changes. It relies on references/project-map.md to locate core directories. Analysis of binary assets is limited to name-string matching.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Code Quality & Review.
Requires project-map.md to function. Cannot delete files or perform automated refactors. Binary asset analysis is limited to name-based string matching in .uasset files.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 9 days ago
- Software
Git