aufgaard

Show current production AI cost overview — totals, top spenders by agent, per-provider breakdown, anomaly detection, and time-to-429 rate-limit prediction. Use when the user asks "what's my AI bill", "which agent is burning tokens", "am I about to hit the rate limit", or wants a cost dashboard. Cross-provider — works for Anthropic, OpenAI, Gemini, Bedrock, Ollama.

Scan Python source files for swallowed-exception patterns that silently turn errors into fake successes. Catches bare `except` blocks that pass / return None / return mock objects, log-and-fake-success handlers, and mock-substitution-on-error. AST-based — not just regex. Use before any deploy of new agent code, on the working directory after a bug fix, or routinely on production-path Python files.

Show current AI deployment health overview — gateway status, plugin/skill registry, recent errors, CPU/RAM pressure, OOM history, cron status, disk pressure, upgrade outcome. The "vital signs" panel for a production AI deployment. Use when the user asks "is my deployment healthy", "what's wrong with my agent", "are services up", or wants an at-a-glance status check.

One-page operator cheatsheet for the Aufgaard plugin. The "what to check, when, why" reference. Auto-loaded so day-to-day routing is fast. Use as the lightweight summary when the user wants a quick recap of capabilities.

Comprehensive production-AI deployment audit against the 35-pattern catalogue. Calls all 7 MCP servers in parallel — bash-vet, skill-vetter, cost-tracker, silentwatch, health-mcp, upgrade-orch, output-vetter — and synthesizes a one-page report with critical findings, audit score, and remediation recommendations. Use when the user asks "is my AI deployment healthy?", "audit my production AI setup", "run a production audit", or wants an end-to-end review of their AI deployment.

Check whether upgrading a package / runtime / model version is safe. Looks up the user-driven regression catalogue (8+ entries from real field reports) AND runs provider-side regression detection (catches Anthropic-April-23-style silent reasoning-effort downgrades). Returns a recommended upgrade path with mitigations. Use before any significant package upgrade, model-version bump, or runtime change.

Show recent silent-failure detections from cron / scheduled jobs. Catches the textbook patterns — exit-0 with empty stdout, length anomalies (output dramatically shorter than baseline), retry storms, action-budget leaks. Use when the user asks "is anything silently broken", "did Friday's cron actually run", "are my scheduled jobs working", or after a downstream consumer reports stale data.

Verify whether an agent's stated outcome ("I committed and pushed", "tests pass", "I cleaned up the temp dir", "deployment succeeded") matches actual filesystem / git / test state. Catches the chiefofautism failure mode (agent confidently misreports what it did) AND the Codex sandbox-escalation case (agent acknowledges read-only constraint then violates it). Use when you suspect an agent's completion claim doesn't match reality, or as a routine post-action check on any state-modifying tool call.

Vet a shell command for production safety BEFORE running it. Catches destructive patterns — rm -rf with unset vars, glob wipeouts, dd/mkfs filesystem destruction, base64-pipe-shell exfil obfuscation, chmod 777 / privilege escalation, force-push, reset --hard. Returns ALLOW/WARN/BLOCK with rule citation. Use when the user pastes a command they're unsure about, OR when reviewing a chain of commands an agent just emitted, OR before approving any shell action with destructive verbs.

Vet an agent-config file or directory (CLAUDE.md, AGENTS.md, .cursor/rules.md, .gemini/config, .claude/skills/, .git/hooks/) BEFORE the agent reads it on next session-start. Catches the agent-config-trust-boundary attack class — adversary lands a config file in a PR, agent inherits the override, RCE-equivalent. 24+ rules including PROMPT_INJ, EXFIL, DYNAMIC_EXEC, SECRET_REF, GIT_HOOK_INSTALL. Use when reviewing PRs that touch any agent-config layer, or after pulling a branch that may have modified these files.

Vet a third-party Claude/Cursor/agent skill (or plugin / extension package) BEFORE installing it. Catches malicious payloads — prompt injection patterns, hardcoded webhook exfiltration, encoded payloads, dynamic execution, suspicious dependencies, typosquatted package names. Returns ALLOW/WARN/BLOCK with rule citation. Use when the user is about to install a community skill, when reviewing a PR that adds a third-party plugin, or after seeing a "this skill 10x'd my agent" tweet that looks too good.

Scoring rubric for evaluating production-AI deployment readiness. The marquee `production-audit` skill uses this rubric to compute the 0-100 score. Auto-loaded so scoring is consistent across sessions and so the user can self-score independently. Use this rubric whenever auditing, reviewing, or quantifying a deployment's production-readiness.

Glossary of production-AI failure modes with verbatim operator phrasings. When the user uses any of these phrases or describes a symptom matching them, you'll recognize the pattern and respond with the right framing + MCP routing. Auto-loaded so the buyer-vocabulary is always available.

Quick reference mapping operator symptoms to the right Aufgaard MCP tool call. Auto-loaded so Claude can route queries to the correct MCP without thinking it through from scratch each time. Use this as the first-resort routing logic when a user describes a symptom, asks "what tool should I use", or when production-audit needs to fan out to specific tools.

Reference catalogue of 35 documented production-AI failure patterns. Each pattern has a P-number, name, indicators, root cause, severity, and which Aufgaard MCP catches it. Auto-loaded when discussing reliability, audit, deployment health, specific failure modes, agent safety, or any phrase from the catalogue. Use this skill's content to ground recommendations in named patterns rather than ad-hoc advice.