Agent
Read-only audit agent that checks whether the current work is drifting from the Product Anchor, Engineering Principles, and current-evidence standards. Use when work must be validated against the product promise, intended experience, anti-shortcut rules, or freshness requirements.
Agent
Prompt engineering specialist that creates or refines prompt artifacts using the embedded Prompt Engineering Bible. Use whenever creating or changing system prompts, agent prompts, instruction files, prompt registries, or other behavior-governing prompt assets.
Agent
Same-tree read-only red-team auditor for active VibeOS sessions with uncommitted code.
Agent
Adversarial agent that hunts for ways tests could pass while software is broken. Finds cheating patterns, mock-reality divergence, vacuous assertions, and status inflation. Reports a corruption score.
Agent
Same-tree read-only security auditor for active VibeOS sessions with uncommitted code.
Agent
Isolated security audit agent that performs OWASP Top 10 analysis, secrets detection, injection pattern scanning, and PII exposure checks. Returns structured findings with severity and confidence ratings.
Agent
Same-worktree System Invariant Auditor for fast session-scoped review of state rules, ownership guarantees, retries, concurrency, partial failure, and recovery behavior without creating an isolated worktree.
Agent
Read-only auditor that validates system invariants: rules that must always remain true across state transitions, auth boundaries, data ownership, retries, concurrency, partial failure, recovery, and future change. Use for Comp outputs, enterprise MVP foundations, critical workflows, data models, auth-sensitive…
Agent
Same-tree read-only test auditor for active VibeOS sessions with uncommitted code.
Agent
Isolated test quality audit agent that evaluates whether tests were written from spec, detects fallback masking, checks mock density, verifies edge coverage, and analyzes git history for TDD compliance.
Agent
TDD agent that reads a WO spec and acceptance criteria, then writes tests BEFORE implementation exists. Never reads implementation code. Tests define expected behavior from the spec.
Hook
Runs when a session starts on startup, executing prereq-check.sh. From chieflatif/vibeos-plugin.
Hook
Runs when you submit a prompt, before the agent sees it, executing governance-guard.sh and intent-router.sh (2 commands). From chieflatif/vibeos-plugin.
Hook
Runs before the agent uses a tool for Bash, Edit and Write tool calls, executing worktree-bash-guard.sh, secrets-scan.sh, frozen-files.sh, test-file-protection.sh, test-diff-audit.sh, worktree-scope-guard.sh, proof-protection.sh and file-budget.sh (8 commands). From chieflatif/vibeos-plugin.
Hook
Runs after a tool call finishes for Bash tool calls, executing gate-result-capture.sh. From chieflatif/vibeos-plugin.
Hook
Runs when a subagent finishes, executing validate-agent-return.sh. From chieflatif/vibeos-plugin.
Hook
Runs before the context is compacted, executing state-flush.sh. From chieflatif/vibeos-plugin.
Hook
Runs when a session ends, executing state-flush.sh. From chieflatif/vibeos-plugin.
Hook
Runs when the agent sends a notification, executing limit-warning-capture.sh. From chieflatif/vibeos-plugin.
Hook
Runs when a git worktree is created, executing worktree-scope-setup.sh. From chieflatif/vibeos-plugin.
Hook
Runs on the TaskCreated event, executing team-governance.sh. From chieflatif/vibeos-plugin.
Hook
Runs when a task is marked completed, executing team-governance.sh. From chieflatif/vibeos-plugin.
Hook
Runs on the TeammateIdle event, executing team-governance.sh. From chieflatif/vibeos-plugin.
Hook
Runs on the ConfigChange event, executing model-policy-guard.sh. From chieflatif/vibeos-plugin.