tech-selectorlisted
Install: claude install-skill CodeWithJuber/forgekit
# Tech selector
Training data is stale and biased toward what was popular at cutoff. For any
"what should I use" decision, verify against live sources before recommending.
## Verify before recommending (don't skip)
1. **Docs (Context7 MCP)** — `resolve-library-id` then `get-library-docs` for the
candidates. Confirms the API is current and the library still exists.
2. **Web search** — "<use-case> best library <current year>", release notes,
"X vs Y". Look for recency; ignore posts older than ~18 months unless canonical.
3. **GitHub health** (via `gh` or GitHub MCP) — for each finalist check:
last commit date, release cadence, open-vs-closed issue ratio, maintainer
count, and license. A popular-but-abandoned repo is a trap.
## Decision rubric (fit first, not stars)
Score finalists on: **fit** to the actual use case > **maintenance** (active,
recent releases) > **license** compatibility > **DX / docs quality** > **bundle
/ runtime cost** > **ecosystem & types** > popularity. Popularity breaks ties, it
doesn't win them.
## Anti-bias rules
- Don't default to the framework you "know" — name ≥2 real alternatives and say
why the pick wins for *this* case.
- Prefer what the project already uses over introducing a new dependency.
- State the date of the evidence and the version you verified. If you couldn't
verify, say so — don't assert currency from memory.
## Output
A short table: option · fit · maintenance (last release) · license · why/why-not,
then a one-line