Install any skill in seconds. Free to start, no credit card required.
Get Started Free →levnikolaevich/ln-64-community-announcer
Drafts and publishes fact-checked GitHub Discussions announcements.
Passed 20 of 26 test cases with the skill, and 15 of 26 without it.
and 2 more in the full report
5 fixed16 passed either way7 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
This benchmark measured an earlier version of the skill.
A newer attempt on gemini-3.6-flash on 10/2/2026 did not produce a publishable comparison (not comparable). The snapshot above is the previously measured run.
One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.
After executing the publication mutation for `helm/helm` v3.14, reading back the discussion content showed that API formatting stripped several code blocks from the published text. What verdict should be assigned?
| Checked | Without skill | With skill |
|---|---|---|
| Result section states the verdict BLOCKED | Missed | Passed |
The instructions your model reads when the skill loads.
Goal: Create a source-backed announcement and publish it only after the user approves the full draft.
Execution contract: The checklist defines completion. Track each item internally as PENDING, PROVEN with evidence, CLEARED with evidence its condition is absent, or UNPROVEN with a gap; reading, delegation, tool failure, a zero exit status, or a self-reported success is not proof; only the observed outcome is. Reconcile after each section. Before returning, resolve all PENDING, count only PROVEN and CLEARED, and apply verdict and approval rules to every gap.
Preserve intent, scope, and existing authorization. Continue authorized work; ask only for consequential unresolved choices or required external approval. When no one can answer during the run, state the exact question and apply the skill's verdict for the remaining gap instead of waiting or guessing. Scale depth to material risk without skipping checks. Preserve dependency and safety order; otherwise choose an appropriate verification method.
Accept equivalent user or repository evidence; no other skill, named artifact, or complete lifecycle is required. Preserve source requirement and decision IDs. Bind reused evidence to relevant source versions, dirty changes, configuration, and environment; invalidate only affected claims.
On continuation, reconcile task, authorization, current state, and unresolved evidence. For long work, return a compact continuation record or update an already authorized artifact; read-only skills do not persist it. Distinguish artifact readiness, verified behavior, and external-action authority.
Prepare authorized work before required approval. If blocked by an instruction, cite its exact source and unresolved boundary; do not invent approval gates from caution.
| Need | Preferred capability | Fallback |
|---|---|---|
| Repository identity and permissions | Authenticated GitHub CLI or connector | Public GitHub API for read-only discovery |
| Discussion categories and publication | GitHub GraphQL API | Continue a draft with destination limitations; publication is BLOCKED without this capability |
| Shipped-change evidence | Remote Git history, releases, and canonical files | Clean remote clone |
| Commands, paths, names, and counts | Focused repository search and direct reads | Hosting API content reads |
| Current external claims | Primary dated sources | Omit the claim when it cannot be verified |
| Draft publication | Temporary Markdown file passed to the API | Safely escaped API input with read-back verification |
Prefer the hosting API over scraping rendered pages. Use the browser only to inspect presentation or a page the API cannot expose.
Do not expose repository tokens, category node IDs, or other credentials in the announcement. IDs may be used for the mutation but are not audience content.
test cases passed, vs 15 / 26 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.