Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →varnan-tech/sdk-adoption-tracker
Given your SDK or library name, searches GitHub code search for public repos that import or require it, classifies each repo as company org, affiliated developer, solo developer, or tutorial noise, scores by adoption signal strength, detects new adopters by date, and outputs a ranked list of who is building on you with outreach context per high-signal company.
Passed 11 of 18 test cases with the skill, and 6 of 18 without it.
and 8 more in the full report
9 fixed4 passed either way6 still failing3 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 9 test cases the skill fixed. Same model, same prompt, with and without the skill.
In the adoption scoring model, what point bonuses are awarded if a repo's `days_since_push` is under 30 days and under 7 days?
| Checked | Without skill | With skill |
|---|---|---|
| Adds 30 points to adoption score for repos pushed within the last 30 days. | Missed | Passed |
| Adds an additional 20 points to adoption score for repos pushed within the last 7 days. | Missed | Passed |
The instructions your model reads when the skill loads.
Take an SDK name. Search GitHub for public repos that import it. Score each repo by company signal, activity, and noise indicators. Enrich high-signal repos with owner and contributor data. Output a ranked adoption report with outreach context for company adopters.
Critical rule: Every repo in the output must exist in the GitHub code search API response. Every company name must come from the GitHub user or org API company or name field. Every contributor handle must come from the GitHub contributors API response. If any field is empty in the API, write "not listed" -- do not infer, guess, or extrapolate.
| The agent will want to... | Why that's wrong |
|---|---|
| Run code search without GITHUB_TOKEN | Unauthenticated code search hits a 3 req/min secondary rate limit and fails on any meaningful scan. GITHUB_TOKEN is required. Stop at Step 1 with a clear error if it is missing. |
| Include forks of the SDK itself | Repos that fork the SDK are contributors or mirrors, not adopters. Filter out repos where fork == true AND the repo name matches the SDK name. |
| Send all 500 raw search results to the AI | Code search can return up to 500 results, most of which are noise. Filter and score locally first. Send only the top 20 high-signal repos to the AI analysis step. |
| Report tutorial and example repos as adopters | Repos with "example", "tutorial", "demo", "learn", "sample", "playground", "starter" in the name or description are not production users. Mark as tutorial_noise and exclude from lead briefs. |
| Invent company names or contact handles | Every company name must come from the GitHub company or org name field. Every contributor handle must come from the contributors API response. If a field is empty, write "not listed". |
| Use one import pattern for all ecosystems | require('sdk') will not find Python users. Auto-detect ecosystem from the SDK name and build ecosystem-specific patterns. Ask the user if auto-detection is ambiguous. |
if [ -z "$GITHUB_TOKEN" ]; then
echo "ERROR: GITHUB_TOKEN is required for code search."
echo "Add a token at github.com/settings/tokens (no scopes needed for public repos)."
echo "Without it, GitHub code search hits a 3 req/min secondary rate limit and fails."
exit 1
fi
echo "GITHUB_TOKEN: set"
curl -s -H "Authorization: Bearer $GITHUB_TOKEN" \
-H "Accept: application/vnd.github+json" \
"https://api.github.com/rate_limit" | python3 -c "
import json, sys
d = json.load(sys.stdin)
search = d['resources']['search']
core = d['resources']['core']
test cases passed, vs 6 / 18 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.