Use before claiming Codewhale release work is done: a pointer to docs/RELEASE_CHECKLIST.md for the gates and to cw-dogfood for the manual QA targets.
75%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Low
Low-risk findings worth noting
Fix and improve this skill with Tessl
tessl review fix ./docs/skills/codew-release-qa-sweep/SKILL.mdThe skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
The required workflow reads first-party docs (docs/RELEASE_CHECKLIST.md, RELEASE_RUNBOOK.md, cw-dogfood) but cites GitHub issue numbers (#3216, #2211, #3289, #3227, #3203) as the motivating reports, which an agent may need to look up, and those issues are outsider-authored text it would select and fetch.
Hmbown/Codewhale
3e1359f
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.