Append a tech debt entry to a project tech-debt log (docs/tech-debt-log.md) based on the current conversation. Use when: (1) the user says 'add tech debt', 'log tech debt', 'log this as tech debt', 'note this for the tech debt log', '/add-tech-debt', (2) the user wants to record an architectural smell, coding-practice issue, or non-blocking design flaw discovered while working on something else, (3) the user describes a self-healing or workaround-able issue that signals a deeper design mistake worth logging. Triggers on tech debt logging, architectural smell capture, code quality issue recording.
68
82%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Passed
No findings from the security scan
Append a numbered entry to the tech debt log based on what was discovered in the current conversation.
The log lives at docs/tech-debt-log.md in the repo root (adjust to your project's docs location). It captures architectural smells and "this wasn't thought through" patterns surfaced while working on something else. Entries are not blockers — they're things worth fixing when the area is revisited, or worth knowing about when onboarding.
Look back at the conversation. The tech debt should be something concrete that was investigated, debugged, or noticed in passing — not a hypothetical concern.
A good tech debt entry has:
If the conversation does not contain enough specifics to fill out all five fields below, ask the user one focused question to fill the gap. Do not invent file paths or fabricate severity.
Read docs/tech-debt-log.md to:
## N. headings)Use this exact structure. Match the established voice: blunt, observation-first, no corporate softening. Lowercase file paths. No em-dashes in prose (use regular hyphens or rewrite). No marketing-style adjectives.
## N. <short title, sentence case, no trailing period>
**Where**
- `relative/path/to/file.ts:LINE` (function or symbol name in parens if useful)
- additional locations, one per line
**What's wrong**
One or two short paragraphs. Describe the actual mechanism that's broken or sloppy. Be specific about what the code does versus what it should do. Skip vague language like "could be improved".
**Why it matters**
One paragraph. Name the architectural smell: cross-service coupling no one documented, missing invariant, leaky abstraction, hidden coordination, etc. If a real user lost time to this, say so. The goal is to make the future reader understand the *category* of mistake, not just this instance.
**Right shape**
Bulleted list of plausible fixes. Order from cheapest to most-correct. Each bullet is one sentence. Do not commit to a choice unless the conversation already decided one.
**Severity**
Low / Medium / High. One sentence on blast radius and self-heal behavior.
**Discovered**
YYYY-MM-DD, brief context: what was the user actually doing when this surfaced.
---Use Edit to append the new entry to the bottom of tech-debt-log.md. Keep the trailing --- separator so the next entry has a clean boundary. The numbering is global and monotonic — never reuse a number even if older entries get resolved (resolved entries should be marked, not renumbered).
Report: entry number, title, and one-line summary. Do not paste the whole entry back unless asked. If the entry was a duplicate of an existing one and you updated it instead, say so.
team-conventions.md, not here.d58ca85
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.