菜单
Shared operational guide for RCA agents (orchestrator and critic): data sources, tooling quickstart, service profile protocol, tool discipline, and anomaly definition.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
Structured procedure to locate the root cause of a confirmed bug. Uses keyword search → focused reading → call-graph tracing. Outputs a specific file + line + hypothesis.
Step-by-step procedure to reproduce a bug. Use this FIRST before trying to fix anything. Produces a minimal script that demonstrates the failure.
AgentM SDK development guide — manifest-as-agent-unit philosophy, SDK programmatic invocation, dynamic workflow orchestration, atom contract, Operations abstraction, event system, service communication, CLI conventions, scenario authoring, logging, structured output, and config resolution. Use whenever writing, editing, or reviewing code under src/agentm/ (atoms, core, gateway), contrib/scenarios/ (manifests), or contrib/extensions/ (workspace-member atoms). Also trigger when creating new atoms, modifying MANIFEST declarations, registering tools or events, touching FileOperations / BashOperations / ResourceWriter, writing CLI subcommands, configuring model profiles, spawning child sessions, building multi-agent orchestrators, writing workflow scripts, or when a code change looks like it might bypass the SDK's existing abstractions. If you catch yourself about to write raw os.stat / open() / subprocess.run in an atom, importing openai/anthropic directly instead of going through the provider layer, or shelling
Guide for adding new benchmarks to the agentm-eval framework and running experiments through the unified `agentm eval` CLI. Covers the adapter interface, experiment lifecycle, result schema, ClickHouse linking, and output conventions. Use whenever creating a new eval adapter, modifying an existing benchmark integration, writing code under contrib/evals/src/agentm_eval/, discussing experiment management, or when the user mentions "add a benchmark", "new eval", "接入评估", "新benchmark", "实验管理", "eval adapter". Also trigger when reviewing eval-related code or debugging experiment output structure.
How to write workflow scripts that orchestrate multiple agents in AgentM. Covers agent(), parallel(), pipeline(), phase(), return values, atom_config data passing, and the journal-based resume / invalidation / partial re-run mechanism (workflow_lineage, workflow_invalidate). Trigger when writing a workflow script, choosing an orchestration pattern, recovering from a wrong node result, or when the user says 编排 / workflow / 多 agent / 并行 / 扇出 / 调度 / 写个 workflow / 失效 / 部分重跑 / 重算. Also trigger when you catch yourself building prompts inside a workflow script — pass data via atom_config instead and let the agent build its own prompt.
Analyze agentm session traces using `agentm trace` atomic commands. Use when inspecting trajectories, token economics, tool calls, or aggregating stats across a trace tree (parent + child sessions). Compose atomic commands via shell pipes — never parse OTLP JSONL directly.
基于 SOC 职业分类
| confidence | fact |
| description | Shared operational guide for RCA agents (orchestrator and critic): data sources, tooling quickstart, service profile protocol, tool discipline, and anomaly definition. |
| name | rca-worker-guide |
| tags | ["rca","worker","operational"] |
| type | skill |
Operational knowledge shared by all RCA worker agents. Load this before your first data query.
Two observability periods are available:
abnormal_* tables): the incident windownormal_* tables): the baseline windowEach period has logs, traces, and metrics. An anomaly only exists if there is a significant delta between the two periods.
describe_tables to see all available tables and columns.diagnose-sql skill (loaded alongside this guide) for query rules (column quoting, duration units, correct column names) and the per-signal recipes — the entry SKILL.md plus the same-directory traces.md / metrics.md / logs.md / correlation.md references. Skipping this causes query errors.abnormal_* (incident) and normal_* (baseline).query_sql. Always include LIMIT.think tool is always available for structured reasoning between queries.You have access to a shared Service Profile store — use it to avoid redundant work.
query_service_profile(service_name="X") to check existing findings. Focus on what's MISSING.update_service_profile(...) to record discoveries. Other parallel agents see your updates immediately.anomaly_summary: one terse line (e.g., "p99 60s vs 4s, 45% errors")key_observation: one factual sentence with the metric/field name used.
You have a budget of tool calls for this task. Use them wisely.
think is FREE — does not count against budget. Use it liberally.think to update your mental model and decide the next step.normal_* to quantify the deviation.args, config, or nested objects.A true anomaly MUST appear in abnormal data but be absent (or negligible) in normal data. Always cross-check against normal-period data before reporting anything as anomalous.
Patterns present in BOTH periods are baseline behavior — exclude them. Healthy services are invisible: if a service shows no significant deviation, do not mention it.