From engineering-workflow
Use this skill when the user wants to audit code consistency issues — when the same problem is solved in different ways across the codebase. Activate on /realign.
How this skill is triggered — by the user, by Claude, or both
Slash command
/engineering-workflow:realignThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Activate when the user invokes `/realign`.
Activate when the user invokes /realign.
The user provides two arguments:
"user group assignment", "volume migration", "error handling")Example: /realign src/services/payments "error handling"
Read the scope path. Use the hint to identify the specific pattern or operation the user noticed. Understand how the anchor implements it.
Search beyond the scope for every other place in the codebase that solves the same or related problem. The scope is where the user noticed it — not a fence around the search.
Look for two categories of inconsistency only:
fetch vs axios, declarative config vs imperative script)Do NOT flag:
Group findings by pattern. For each group:
[file:line](path#Lline) format so the user can jump to surrounding context in their editorExample group:
## Error handling in service calls
Found 3 different approaches:
**Try/catch with logger** — [src/services/payments/charge.ts:42](src/services/payments/charge.ts#L42)
\```ts
try {
await stripe.charge(amount);
} catch (err) {
logger.error(err);
throw err;
}
\```
**Result type** — [src/services/payments/refund.ts:18](src/services/payments/refund.ts#L18)
\```ts
const result = await safeCall(() => stripe.refund(chargeId));
if (!result.ok) return { error: result.error };
\```
**Unhandled** — [src/services/payments/webhook.ts](src/services/payments/webhook.ts)
(no error handling found around external call at line 31)
For each pattern group, propose a canonical form — the clearest, most explicit implementation found in the repo, or your own recommendation if none stands out. State your reasoning in one sentence.
Ask the user which groups they want to standardize and whether they agree with the proposed canonical. Continue conversationally from there — no menus or numbered choices.
Guides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.
npx claudepluginhub sliekens/agentic --plugin engineering-workflow