From supervibe
Use WHEN audit flagged weak/stale artifacts to deepen them from project context AND fresh research (consults research-agents for stale best-practices/dependency/security/infra/design). Triggers: 'усиль агента', 'strengthen <agent>', 'докрути артефакт', 'усилить'.
npx claudepluginhub vtrka/supervibe --plugin supervibeThis skill is limited to using the following tools:
WHEN `supervibe:audit` flagged weak or stale artifacts, OR user runs `/supervibe-strengthen`. Loops over each flagged artifact.
Mandates invoking relevant skills via tools before any response in coding sessions. Covers access, priorities, and adaptations for Claude Code, Copilot CLI, Gemini CLI.
Share bugs, ideas, or general feedback.
WHEN supervibe:audit flagged weak or stale artifacts, OR user runs /supervibe-strengthen. Loops over each flagged artifact.
Lifecycle: intake -> plan -> review -> approved -> applied -> verified. Persist state in .supervibe/memory/strengthen/state.json before every lifecycle transition.
Every interactive step asks one question at a time using Step N/M or Step N/M. Each question lists the recommended/default option first, gives a one-line tradeoff summary for every option, allows a free-form answer, and names the stop condition.
Default behavior: produce a dry-run diff and do not edit artifacts until approval. Free-form path: the user can name exact artifacts, agents, rules, or research constraints to include or exclude.
After every material delivery, ask one explicit next-step question about the strengthening diff. Use buildPostDeliveryQuestion({ intent: "strengthening_delivery" }, { locale }) when tooling is available. Visible labels must be language-matched and domain-specific; keep internal action ids only in saved state. Never show both English and Russian in the same visible option.
English visible labels:
Russian visible labels:
Follow docs/references/skill-expert-operating-standard.md: start from source of truth, preserve retrieval evidence, apply scope safety, use real producers with runtime receipts for durable delegated outputs, verify before completion claims, and keep confidence below gate when evidence is partial.
.supervibe/confidence-log.jsonl for override patterns.supervibe/memory/effectiveness.jsonl for agent failure patternsnode <resolved-supervibe-plugin-root>/scripts/build-code-index.mjs --root . --list-missing, then node <resolved-supervibe-plugin-root>/scripts/build-code-index.mjs --root . --resume --source-only --max-files 200 --max-seconds 120 --health --json-progress for source-readiness repair or node <resolved-supervibe-plugin-root>/scripts/build-code-index.mjs --root . --resume --graph --max-files 200 --health for graph repair, and re-check with node <resolved-supervibe-plugin-root>/scripts/supervibe-status.mjs --index-health --no-gc-hintsnode <resolved-supervibe-plugin-root>/scripts/supervibe-status.mjs --capabilities before editing agents, rules, commands or skills so every strengthened artifact stays linked to a capability and verification hook.Artifact is stale (last-verified >90d)?
├─ YES:
│ ├─ References "best practices" / current patterns?
│ │ → MUST invoke supervibe:_ops:best-practices-researcher first
│ ├─ References dependencies / library versions?
│ │ → MUST invoke supervibe:_ops:dependency-researcher first
│ ├─ References security patterns / CVEs?
│ │ → MUST invoke supervibe:_ops:security-researcher first
│ ├─ References infrastructure topology?
│ │ → MUST invoke supervibe:_ops:infra-pattern-researcher first
│ ├─ References competitive design / UX patterns?
│ │ → MUST invoke supervibe:_ops:competitive-design-researcher first
│ └─ Otherwise → strengthen from project context only
└─ NO (just weak, not stale): strengthen from project context only
version (1.0 → 1.1)
h. Update last-verified to today
i. Update verified-against (current commit hash)
j. If researcher consulted: cite source in artifact footernode <resolved-supervibe-plugin-root>/scripts/build-code-index.mjs --root . --resume --source-only --max-files 200 --max-seconds 120 --health --json-progress; use --force --health only for a deliberate full rebuild
b. Confirm SUPERVIBE_INDEX_GATE READY: true
c. If generated leakage or stale rows remain, inspect --explain-policy before editing source rulesReturns:
supervibe:audit — produces input listagents/_ops/*-researcher — consulted for stale artifacts