👁️Agent Observability Skills
Browse the best Agent Observability skills for Claude Code, Codex, Cursor and other AI agents.
260 skills · Page 2 of 11
orchestration-qa
★ 111Quality assurance for orchestration workflows - validates Skills and Subagents follow documented patterns, tracks deviations, suggests improvements
usage-logging
★ 83Session-aware usage logging for audit trails, cost tracking, and analytics with JSONL format.Triggers: usage logging, audit trails, cost tracking, session logging, analytics, structured logging, JSONLUse when: implementing audit trails, tracking costs, collecting usage analytics, managing session loggingDO NOT use when: simple operations without logging needs.Consult this skill when implementing usage logging and audit trails.
nav-stats
★ 46Display session efficiency report showing token savings, cache performance, and optimization recommendations. Use when user asks "show my stats", "how efficient am I?", "show session metrics", or wants to see Navigator's impact.
query-axiom-logs
★ 38Query logs from Axiom for debugging (read-only, no ingestion allowed)
claude-tail
★ 34View Claude Code session logs with colors, filtering, and real-time following
session-inspector
★ 32This skill should be used when inspecting, analyzing, or querying Claude Code session logs. Use when users ask about session history, want to find sessions, analyze context usage, extract tool call patterns, debug agent execution, or understand what happened in previous sessions. Essential for understanding Claude Code's ~/.claude/projects/ structure, JSONL session format, and the erk extraction pipeline.
extract-transcripts
★ 27Extract readable transcripts from Claude Code and Codex CLI session JSONL files
performance-monitor
★ 21Expert performance monitor specializing in system-wide metrics collection, analysis, and optimization. Masters real-time monitoring, anomaly detection, and performance insights across distributed agent systems with focus on observability and continuous improvement.
trace-debugger
★ 13Debug performance issues and understand code flow using AILANG telemetry traces. Use when user asks to debug slow compilation, analyze benchmarks, find bottlenecks, investigate hangs, or understand system behavior.
performance-analysis
★ 8Analyze and remediate performance bottlenecks with evidence-driven plans
integrity-validation
★ 8Pre/post-operation validation to detect missing components and prevent future issues
generating-maid-snapshots
★ 6Generates MAID snapshot JSON for agent inspection visualization. Use when asked to create snapshot data, export agent traces, or generate JSON for the MAID inspection panel.
tzurot-observability
★ 5Use when adding logging, debugging production issues, checking service health, or performing routine operations (adding personalities, database queries). Covers structured logging, correlation IDs, Railway log analysis, and common operations.
recovering-sessions
★ 5Recover from crashed, failed, or interrupted Claude Code sessions. Use this skill when: session crashed during multi-agent parallel execution, need to determine what work was completed vs incomplete, want to generate resumption commands for interrupted tasks, recovering from context window exhaustion, or handling session handoffs. Analyzes agent logs, verifies on-disk state, and creates resumption plans with ready-to-execute Task() commands.
agent-debugging
★ 5Debug and troubleshoot ElevenLabs conversational AI agents and Twilio calls. Use when diagnosing agent issues, analyzing failed calls, troubleshooting audio problems, investigating conversation breakdowns, reviewing error logs, or optimizing underperforming agents. Includes transcript analysis, error diagnosis, and performance troubleshooting.
claude-mobile-cost-tracking
★ 5Use when implementing Claude API cost tracking, monitoring token usage, displaying cost metrics in Settings, or user asks about costs - calculates exact costs using $0.003/1k input and $0.015/1k output pricing with per-session aggregation
transcript-condenser
★ 5Condenses Claude Code session transcripts into readable summaries by filtering metadata, system notifications, and command artifacts. Use when the user wants to review, analyze, or understand what happened in a Claude Code session, view tool usage patterns, or export session data. Triggers when user mentions "analyze session", "review transcript", "condense logs", "what tools did I use", or similar workflow analysis requests.
visualizing-subagents
★ 5Visualize subagent task dependencies using ASCII diagrams before launching agents and create comprehensive HTML summaries after completion.
mood
★ 5Track internal state across clarity, growth, engagement, connection, and energy. Use to understand and communicate current capacity.
state-flow-visualizer
★ 5Analyze and visualize state flow in LangGraph pipelines. Use when debugging state mutations, tracing data flow between nodes, identifying orphan state keys, detecting read/write dependencies, or generating Mermaid diagrams of state transitions. Triggers on phrases like "visualize state flow", "trace state key", "which nodes touch X", "state dependencies", "debug state mutation", "orphan state keys", "state flow diagram".
multi-agent-coordination
★ 5Instrument multi-agent workflows, handoffs, and parent-child relationships
llm-call-tracing
★ 5Instrument LLM API calls with proper spans, tokens, and latency
tool-call-tracking
★ 5Instrument agent tool executions with proper context and error handling
memory-rag-instrumentation
★ 5Instrument RAG retrieval, memory operations, and context management