aisolve

Converts TODO or XXX comments into code fixes via git grep and codepath analysis.

20|1|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/radareorg/radare2-skills --skill aisolve
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aisolve
Source: https://github.com/radareorg/radare2-skills/tree/main/dev/aisolve
Command: npx skills add https://github.com/radareorg/radare2-skills --skill aisolve

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Helps convert ambiguous TODO/XXX comments into concrete, correct code changes by guiding an agent to understand the underlying issue and implement a good solution.

Core Features & Use Cases

  • Finds real TODO/XXX targets: Searches the whole codebase for TODO and XXX markers using git grep, then selects one candidate to tackle.
  • Ensures correctness before coding: Studies the comment’s intent by checking surrounding code, callers, and relevant codepaths to understand the real problem.
  • Produces a reasoned implementation plan: Encourages using explicit reasoning to choose a good approach before writing code, reducing guesswork and regressions.

Quick Start

Use the aisolve skill to find a TODO or XXX comment in the repository and produce a well-reasoned fix with the correct behavior across all impacted codepaths.

Frequently Asked Questions about aisolve

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I convert TODO comments into working code fixes?

Converting TODO comments into working code fixes requires searching the repository via git grep, analyzing the surrounding control flow and callers, planning an approach, and writing the resolved code across all impacted codepaths.

What is the best way to resolve XXX markers in a codebase?

Resolving XXX markers in a codebase is best handled by selecting an actionable target, understanding the underlying issue through whole-repo discovery and static analysis, and validating the implementation across multiple codepaths before applying the fix.

Does resolving inline TODO comments require checking callers?

Yes, resolving inline TODO comments requires checking callers and relevant codepaths to ensure the implementation is validated for correctness and does not introduce regressions across the repository.

How to find real TODO targets across a whole repository?

To find real TODO targets across a whole repository, use git grep to search for TODO and XXX markers, then select one candidate to tackle and analyze its intent by checking the surrounding code.

Why does fixing a TODO comment require a reasoned implementation plan?

Fixing a TODO comment requires a reasoned implementation plan to reduce guesswork and regressions by using explicit reasoning to choose a good approach and validate correct behavior before writing the actual code.

Can I use static analysis to understand ambiguous TODO requirements?

Yes, you can use static analysis to understand ambiguous TODO requirements by studying the comment's intent through surrounding code, callers, and control flow to plan a good solution before writing code.