Collects ponytail: code comments into a technical debt ledger report.
npx skills add https://github.com/dsonyy/dotfiles --skill ponytail-debt-dsonyy
Please help me install this Agent Skill. Skill: ponytail-debt Source: https://github.com/dsonyy/dotfiles/tree/main/config/plugins/ponytail/skills/ponytail-debt Command: npx skills add https://github.com/dsonyy/dotfiles --skill ponytail-debt-dsonyy
ponytail: comments scatter across a codebase and quietly rot into permanent debt. This Skill greps the repository for every marker and compiles them into a single ledger so deferred work stays visible instead of being forgotten. ## Core Features & Use Cases - Marker Harvesting: Greps the repo for ponytail: comment markers while skipping node_modules, .git, and build output. - Structured Ledger: Outputs one row per marker grouped by file, extracting the ceiling (the limit) and the upgrade trigger from each comment, with optional git blame ownership. - Rot Detection: Flags any marker missing an upgrade path with a no-trigger tag so silent debt is surfaced. - Use Case: Before a sprint planning session, run the ledger to list every deferred shortcut, see which ones have no revisit trigger, and decide what to pay down. ## Quick Start Ask the assistant to run ponytail debt and list every shortcut marked in this repository.High-intent search queries and answers about installing and using this skill.
Mark each deliberate shortcut with a `ponytail:` comment naming its ceiling and upgrade path, then run this Skill to grep all markers into one ledger. Each row shows the file, line, what was simplified, and the trigger to revisit it.
Run the ponytail debt scan, which greps the repo for `ponytail:` markers and flags any comment missing an upgrade trigger with a `no-trigger` tag. Those flagged rows are the deferrals most likely to silently rot.
No, the scan is strictly read-only and changes nothing in the repository. It only reports the ledger, and writes it to a file like PONYTAIL-DEBT.md only if you explicitly ask.
It greps for `#` and `//` comment prefixes followed by `ponytail:` by default. If your stack uses other comment syntaxes, you can extend the grep pattern with additional prefixes.
Yes, add `git blame -L<line>,<line>` per row to attach an owner to each ledger entry. This is optional and layered on top of the base grep output.
marmelabScan repository files for ponytail: comment markers and generate a structured debt ledger.
miptah21Scan codebases for technical debt and generate prioritized remediation plans.
XinAlohaScan source files for ponytail comments and generate a debt ledger.
rakhnianskii9Scan codebases for TODO/FIXME/HACK markers and maintain a prioritized debt register.