VibeTags is a Java annotation processor that acts as AI guardrails for code generation tools like Cursor, Claude, Gemini, and Codex CLI. It allows developers to control AI behavior through simple annotations, protecting critical code and guiding AI implementations.
Add a new @AI... guardrail annotation to VibeTags — register it in GuardrailAnnotations, then wire it through every platform renderer, GranularRenderer, BuildFingerprint, and AnnotationValidator, and update all four docs. Use when the user says "add annotation", "new @AI annotation", "add a new guardrail annotation"…
Add support for a new AI coding tool, IDE, PR-reviewer, or context-packer to VibeTags — wire a new generated output file through ServiceRegistry, Platform, the renderer registry, and (if granular) GranularRulesWriter. Use when the user says "add platform support", "support ", "generate a new output file for X", "new…
Bump every third-party dependency, plugin and toolchain pin in VibeTags to its latest stable release, mirror the pins Gradle cannot inherit, verify with the real gates, and open the PR. Use when the user says "bump dependencies", "update dependencies", "dependency update", "upgrade deps", "are we on the latest…
Pre-push adversarial interrogation of the current branch's diff by a fresh reviewer with no memory of writing it - five scoped questions (done? production-ready? scales? secure? top risks?), answers must cite a line or admit "cannot tell from here". Advisory only, never a gate. Use when the user says "consult"…
Build every downstream consumer of VibeTags against a chosen VibeTags version and report which ones actually pass. Use when the user says "regression suite", "consumer sweep", "test the consumers", "check the downstream repos", "does the new version break anything", or before cutting a VibeTags release.
Run a bug hunt against this repository under a fixed work order - baseline the suite, hunt with named bug-shape lenses, failing-test-first, one commit per bug, suspicions reported separately, locks escalated never overridden. Use when the user says "correctness hunt", "bug hunt", "find bugs", "hunt for bugs", or asks…
Run the VibeTags benchmark harness in load-tests/ and report the numbers honestly — annotation-volume and memory sweeps, JMH hot-path and cache-hit microbenchmarks, the processor-tax control — then record a release baseline under load-tests/results/. Use when the user says "load tests", "benchmarks", "capture a…
Cut a new VibeTags release — bump the version everywhere, update the CHANGELOG, open the release PR, and create the GitHub release that triggers Maven Central publishing. Use when the user says "release", "/release", "cut a release", "publish a new version", or asks to bump VibeTags to a new version.
This skill should be used when the user asks how to "use VibeTags", "add VibeTags annotations", "set up AI guardrails", "protect code from AI", "configure AI platforms", asks about @AILocked, @AIContext, @AIDraft, @AIAudit, @AIIgnore, @AIPrivacy, @AICore, @AIPerformance, @AIContract, @AITestDriven, @AIThreadSafe…
AI guardrails and coding rules generated by VibeTags from @AILocked, @AIContext, and related annotations. Load this skill when writing, reviewing, or modifying code in this project.
★not rated 15▲
+1 yesterdayA46 tokens
originalMIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: