Skills
Agent skills published to the LLMBrain gallery — reviewed by hand, installable into any project with one command. Your agent picks them up wherever you work, not just on the machine you downloaded them to.
7 published
coderabbit
Run a CodeRabbit CLI review over the current changes and triage the findings — a deterministic static-analysis and security pass to complement an LLM diff review. Auto-invoke when the user says "code…
1 file · updated Aug 21, 2026andrei-firoiupublic-changelog
Draft a user-facing release note for a PR and append it to public-changelog.md, dated and referencing the PR id. Focus only on functional changes real users can see — skip internal-only work. Auto-in…
1 file · updated Aug 21, 2026andrei-firoiuship-it
Finalize a change end-to-end — adversarial review of the whole feature, format, static analysis, tests, records, a commit in the project's style, and open a PR. Auto-invoke when the user says "ship i…
1 file · updated Aug 21, 2026ghalexgrill-me
Interview the user relentlessly about a plan or design. Use when the user wants to stress-test a plan before building.
1 file · updated Aug 18, 2026ghalexreact-rules
React architecture rules for both auditing and development. Use when asked to "review components", "check architecture", "audit this react project", "does this follow react rules", "review my fronten…
3 files · updated Aug 18, 2026ghalexwork-on-issue
Pick an issue from the brain, ground it against the actual code, and confirm before starting work. Use when the user wants to start working on an issue, asks "what should I work on", or invokes /work…
1 file · updated Aug 20, 2026ghalexwork-on-milestone
Drive a whole milestone to completion as a ralph loop — pick an issue, hand it to a separate worker agent, commit the slice, repeat until the milestone is empty. Use when the user invokes /work-on-mi…
1 file · updated Aug 22, 2026