From jewzaam-reviews
Permission canary — exercises every !-injection pattern used by real skills to verify allowed-tools coverage
How this skill is triggered — by the user, by Claude, or both
Slash command
/jewzaam-reviews:testThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
This skill exercises every `!` injection pattern used by the real producer skills. If any section below shows an error instead of output, the `allowed-tools` pattern is broken.
This skill exercises every ! injection pattern used by the real producer skills. If any section below shows an error instead of output, the allowed-tools pattern is broken.
CLAUDE_PLUGIN_ROOT=${CLAUDE_PLUGIN_ROOT} CLAUDE_PLUGIN_DATA=${CLAUDE_PLUGIN_DATA} CLAUDE_SKILL_DIR=${CLAUDE_SKILL_DIR}
!bash ${CLAUDE_PLUGIN_ROOT}/scripts/print-plugin-home.sh ${CLAUDE_PLUGIN_ROOT}
!pwd
!git remote -v || true
!bash ${CLAUDE_PLUGIN_ROOT}/scripts/bootstrap-tmp.sh .tmp-test
!bash ${CLAUDE_PLUGIN_ROOT}/scripts/print-handoff-contract.sh
Print every section heading above and its injected output. Add a specific output for the plugin and marketplace versions derrived from the auto-injected env var paths. If any section shows "Shell command permission check failed" or similar error, report which section failed and the exact error message. Otherwise confirm all injections succeeded.
npx claudepluginhub jewzaam/jewzaam-reviews --plugin jewzaam-reviewsCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.