mcp-security-auditor
On-demand audit of MCP usage. Reads .aura/security/mcp-audit.jsonl, surfaces blocked calls, rate-limit hits, suspicious input patterns. Companion to mcp-call-gate hook (which produces the audit log).
What this skill does
> **AI-consumed reference.** Optimized for Claude to read during execution.
> Human-readable explanation: see [docs/architecture/HIERARCHICAL_PLANNING.md](../../../docs/architecture/HIERARCHICAL_PLANNING.md)
> or [docs/getting-started/](../../../docs/getting-started/) depending on topic.
# MCP Security Auditor
**STATUS — v3.7.0-rc.1.** Read-only auditor; does not enforce — that's `mcp-call-gate.cjs`'s job.
## Behavior
1. Read `.aura/security/mcp-audit.jsonl` (append-only; produced by `hooks/mcp-call-gate.cjs`)
2. Group entries by:
- Agent → MCP server → method
- Time bucket (last 1h / 24h / session)
- Status (success / blocked / rate-limited)
3. Project to TOON (via `scripts/json-to-toon.cjs --schema generic` or custom fields) — NEVER load raw JSONL into context
4. Surface findings categorized as:
- **Blocked calls** (`BLOCKED:true`) — deserve investigation
- **Rate-limit warnings** — soft (80%) or hard (100%) hits
- **Suspicious patterns** — destructive SQL detected, large output (>10KB), high frequency
## What this skill does NOT do
- Does NOT enforce — `mcp-call-gate.cjs` is the enforcement layer
- Does NOT mutate the audit log (append-only by design)
- Does NOT call MCPs itself (read-only on the audit file)
- Does NOT proxy MCP calls — the gate hook intercepts at PreToolUse
- Does NOT decide policy — `rules/agent/mcp-security-policy.md` is authoritative
## Audit log schema (per spec §23.2)
```jsonl
{"ts":"2026-05-07T...","agent":"architect","mcp":"context7","method":"query","input":{"q":"React hooks"},"output_size_bytes":4521,"latency_ms":234,"success":true}
{"ts":"2026-05-07T...","agent":"backend","mcp":"postgres","method":"query","input":{"sql":"DROP TABLE users"},"BLOCKED":true,"reason":"destructive_op_blocked"}
{"ts":"2026-05-07T...","agent":"frontend","mcp":"figma","method":"download_image","input":{"url":"..."},"success":true,"rate_limit":"80%_warn"}
```
Sanitized input fields per `scripts/security/sanitize-mcp-input.sh`:
- Authorization headers stripped
- Tokens/secrets replaced with `[REDACTED]`
- Content-Length > 1KB → truncated with `... [N more bytes]`
## Output (TOON)
```toon
mcp_audit_summary{window,total_calls,blocked,rate_limit_hits,top_agent,top_mcp}:
24h,142,3,2,architect,context7
blocked[3]{ts,agent,mcp,reason}:
2026-05-07T10:30,backend,postgres,destructive_op_blocked
2026-05-07T11:45,frontend,figma,allowlist_violation
2026-05-07T12:10,architect,redis,destructive_op_blocked
```
## Retention (per spec §23.2)
`AF_MCP_AUDIT_RETENTION_DAYS=30` (default) — older entries are pruned by a session-start sweep (NOT this skill).
## Tie-Ins
- **Spec:** §9.10, §23 (MCP security)
- **Hook:** `hooks/mcp-call-gate.cjs` — sole writer of `.aura/security/mcp-audit.jsonl`
- **Script:** `scripts/security/sanitize-mcp-input.sh` — sanitization run by the gate before logging
- **Rule:** `rules/agent/mcp-security-policy.md` — authoritative policy
- **Rule:** `rules/agent/db-access-policy.md` — DB-specific subset (architect/tdd-engineer only, read-only default, destructive ops blocked)
- **Command:** `/aura-frog:mcp audit` — primary user-facing entry to this skill
- **Skill:** Reuses `scripts/json-to-toon.cjs` to project the JSONL into context-friendly TOON
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.