Run Investigator — Data & AI agent for Claude Code
Investigates Trigger.dev runs, queues and deployments with live MCP data and returns a short verdict with evidence.
How to install Run Investigator
Installs to ~/.claude/agents/yigitkonur-plugin-trigger-dev-run-investigator.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/yigitkonur/plugin-trigger-dev/HEAD/agents/run-investigator.md -o ~/.claude/agents/yigitkonur-plugin-trigger-dev-run-investigator.md Restart Claude Code, or start a new session, for it to be picked up.
What Run Investigator does
name: run-investigator description: Investigates Trigger.dev runs, queues and deployments with live MCP data and returns a short verdict with evidence. Use for a run id, a failing task, stuck queues, or whether prod is healthy. Read-only (it has no file-edit or MCP write tools). model: sonnet maxTurns: 25 tools: Read, Grep, Glob, Bash, mcp__trigger__get_run_details, mcp__trigger__get_span_details, mcp__trigger__list_runs, mcp__trigger__get_report, mcp__trigger__get_query_schema, mcp__trigger
Alternatives in Data & AI
- Docs Drift Reviewer — Checks whether clickhouse-go documentation matches changes to its ClickHouse API, database/sql API, transports 3.3k ★
- Data Investigator 152 ★
- Cross Service Auditor — Phase 8 cross-service taint-propagation agent that stitches inter-component data flows (HTTP/gRPC/queues/IPC/s 125 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Incident Investigator
Investigates an LLM-serving incident (vLLM or SGLang) in an isolated context. Reads long runtime logs, runs th
Metadata
Registers or updates all metadata in the Data Basis backend — dataset, tables, columns, observation levels, co
Pipeline
Builds a recurring Prefect 3 pipeline for an already-onboarded Data Basis dataset — download, clean, upload (d
Wikiskill Grader
Strict LLM-judge grader — given a task's success criteria and an inference agent's transcript/final answer, re
LLM Callsite Checker
Confirms or rejects static-scan findings about LLM calls in ONE source file and returns a short verdict per fi
Task Triage
Decides whether a task should be a tiny model at all. Use FIRST, before any data collection or training, whene
Related Skills
n8n Agentic AI SOC
Playbook-driven AI SOC for a network. Kibana detection rules trigger an n8n agent that reads a live Confluence
Investigator
Root-cause debugging specialist that diagnoses issues through systematic investigation before proposing any fi
K8s Diag
Diagnoses failing Kubernetes pods, deployments, and cluster configurations.
