Monitoring Observability
ahmedasmar/devops-claude-skills
Monitoring and observability strategy, implementation, and troubleshooting.
Quantifies elevated error rates with PromQL, compares them to a baseline, and isolates which jobs or instances an error spike is concentrated in.
Install Prometheus Error Rate Investigator
$ npx skills add prometheus/prometheus-mcp --skill investigate-error-rates -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install prometheus/prometheus-mcp investigate-error-rates --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/prometheus/prometheus-mcp.git skills-src && mkdir -p .claude/skills && cp -r skills-src/pkg/mcp/assets/skills/investigate-error-rates .claude/skills/investigate-error-rates && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "investigate-error-rates" agent skill from https://github.com/prometheus/prometheus-mcp/tree/main/pkg/mcp/assets/skills/investigate-error-rates into .claude/skills/investigate-error-rates/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "investigate-error-rates", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/prometheus/prometheus-mcp/tree/main/pkg/mcp/assets/skills/investigate-error-ratesType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add prometheus/prometheus-mcp --skill investigate-error-rates -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install prometheus/prometheus-mcp investigate-error-rates --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/prometheus/prometheus-mcp.git skills-src && mkdir -p .agents/skills && cp -r skills-src/pkg/mcp/assets/skills/investigate-error-rates .agents/skills/investigate-error-rates && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "investigate-error-rates" agent skill from https://github.com/prometheus/prometheus-mcp/tree/main/pkg/mcp/assets/skills/investigate-error-rates into .agents/skills/investigate-error-rates/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "investigate-error-rates", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add prometheus/prometheus-mcp --skill investigate-error-rates -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install prometheus/prometheus-mcp investigate-error-rates --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/prometheus/prometheus-mcp.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/pkg/mcp/assets/skills/investigate-error-rates .cursor/skills/investigate-error-rates && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "investigate-error-rates" agent skill from https://github.com/prometheus/prometheus-mcp/tree/main/pkg/mcp/assets/skills/investigate-error-rates into .cursor/skills/investigate-error-rates/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "investigate-error-rates", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/prometheus/prometheus-mcp.git --path pkg/mcp/assets/skills/investigate-error-rates--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add prometheus/prometheus-mcp --skill investigate-error-rates -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install prometheus/prometheus-mcp investigate-error-rates --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/prometheus/prometheus-mcp.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/pkg/mcp/assets/skills/investigate-error-rates .gemini/skills/investigate-error-rates && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "investigate-error-rates" agent skill from https://github.com/prometheus/prometheus-mcp/tree/main/pkg/mcp/assets/skills/investigate-error-rates into .gemini/skills/investigate-error-rates/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "investigate-error-rates", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install prometheus/prometheus-mcp investigate-error-ratesInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add prometheus/prometheus-mcp --skill investigate-error-rates -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/prometheus/prometheus-mcp.git skills-src && mkdir -p .github/skills && cp -r skills-src/pkg/mcp/assets/skills/investigate-error-rates .github/skills/investigate-error-rates && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "investigate-error-rates" agent skill from https://github.com/prometheus/prometheus-mcp/tree/main/pkg/mcp/assets/skills/investigate-error-rates into .github/skills/investigate-error-rates/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "investigate-error-rates", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add prometheus/prometheus-mcp --skill investigate-error-rates -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install prometheus/prometheus-mcp investigate-error-rates --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/prometheus/prometheus-mcp.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/pkg/mcp/assets/skills/investigate-error-rates .opencode/skills/investigate-error-rates && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "investigate-error-rates" agent skill from https://github.com/prometheus/prometheus-mcp/tree/main/pkg/mcp/assets/skills/investigate-error-rates into .opencode/skills/investigate-error-rates/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "investigate-error-rates", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
investigate-error-rates- GitHub stars
- 120
- Token cost
- ~592 tokens
- SKILL.md length
- 257 words
- Files
- 1
- Skills in repo
- 7
- Repo updated
- First seen
- Licence
- Apache-2.0
At a glance
Quantifies elevated error rates with PromQL, compares them to a baseline, and isolates which jobs or instances an error spike is concentrated in.
- Investigating a spike in 5xx or gRPC error responses
- SKILL.md covers Getting oriented, Topics worth exploring and Reporting findings
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
- Confirming whether a firing error-rate alert reflects a real traffic-wide problem
What it does
The skill starts by discovering what error signals actually exist in the target system, since they vary: HTTP status counters, gRPC code counters, or dedicated errors or failures metrics, found with series or label-value queries and confirmed against metric metadata, alongside checking list_alerts and the ALERTS firing series for anything already flagged. From there it walks through computing a current error rate with rate() over raw counters, dividing by total traffic to get a ratio that separates more errors from more traffic, and running the same query over a longer range to find the inflection point against a known-good baseline window.
To localize the problem it aggregates the error query by job, instance, handler or method, using topk to keep the output readable, and cross-checks the same time window against deploys or restarts, resource saturation and latency histograms for the same services. The expected output is a clear statement of impact in terms of error ratio and affected traffic, an onset time, and the narrowest scope that explains the spike, backed by the actual queries run.
When your agent uses it
- Investigating a spike in 5xx or gRPC error responses
- Confirming whether a firing error-rate alert reflects a real traffic-wide problem
- Narrowing an error spike down to a specific job, instance or route
Example prompts
- “Our 5xx rate just spiked; find out when it started and which service it is concentrated in.”
- “Check whether this firing alert is a traffic-wide issue or just one instance.”
- “Compare the current gRPC error ratio against last week's baseline for the checkout service.”
Requirements
- A connected Prometheus MCP server
- Compatibility (from SKILL.md): Requires the tools of a connected Prometheus MCP server
What it can do on your machine
Read from SKILL.md and the folder at commit 856fb45. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Compatibility
Requires the tools of a connected Prometheus MCP server
From compatibility in the SKILL.md frontmatter.
Context cost
Prometheus Error Rate Investigator loads about 592 tokens when it runs. Until then it costs about 65 tokens; SKILL.md has 257 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from prometheus/prometheus-mcp at commit 856fb45, republished under its Apache-2.0 licence (© prometheus). 257 words, ~592 tokens.
Download SKILL.mdSave it as
.claude/skills/investigate-error-rates/SKILL.md (or your agent's skills folder).- name
- investigate-error-rates
- description
- Investigate elevated error rates and failing requests. Use when errors or 5xx/gRPC failures spike, SLOs burn, or an error alert fires; quantifies error ratios with rate(), compares to baseline, and isolates affected jobs and instances.
- compatibility
- Requires the tools of a connected Prometheus MCP server
- license
- Apache-2.0
Investigating High Error Response Rates
Figure out how bad the errors are, when they started, and where they are concentrated. Error metrics vary by system, so discover what actually exists before querying.
Getting oriented
- Error signals are usually counters: HTTP status codes (http_requests_total{code=~"5.."}), gRPC codes (grpc_server_handled_total{grpc_code!="OK"}), or dedicated *_errors_total / *_failures_total metrics.
- Discover what this system exposes with series or label_values on name using regex matchers, and confirm semantics with metric_metadata.
- list_alerts shows whether an error-related alert is already firing and carries useful labels to start from.
- The ALERTS{alertstate="firing"} series Prometheus generates is the queryable form of the same information: range_query it to see when alerts started firing and line their history up against the error timeline.
Topics worth exploring
Treat these as starting points and follow what the data shows:
- Current error rate: counters need rate() before aggregating, e.g. with query: sum by (job) (rate(http_requests_total{code=~"5.."}[5m]))
- Error ratio vs raw count: a ratio against total traffic distinguishes "more errors" from "more traffic": sum(rate(http_requests_total{code=~"5.."}[5m])) / sum(rate(http_requests_total[5m]))
- When it started: range_query the same expression over hours or days and look for the inflection point; compare against a known-good baseline window.
- Where it is concentrated: aggregate by job, instance, handler, or method; topk(10, ...) keeps output manageable.
- What else changed at that time: deploys and restarts (process_start_time_seconds), saturation (CPU, memory, connection pools), and latency histograms for the same services.
Reporting findings
Aim for a clear statement of impact (error ratio and affected traffic), onset time, and the narrowest scope (service, instance, or route) that explains the errors, backed by the queries you ran.
© prometheus, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in pkg/mcp/assets/skills/investigate-error-rates of prometheus/prometheus-mcp.
Open the folder on GitHubat commit 856fb45
Compare with similar skills
Prometheus Error Rate Investigator next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Prometheus Error Rate Investigator this skillprometheus/prometheus-mcp | 120 | — | ~592 | Automated safety check: Pass | Apache-2.0 | |
| Monitoring Observabilityahmedasmar/devops-claude-skills | 203 | — | ~3.9k | Automated safety check: Pass | None | |
| Observability MonitoringAnastasiyaW/codex-claude-code-config | 154 | — | ~4.1k | Automated safety check: Pass | MIT | |
| Observability Sremajiayu000/spellbook | 287 | — | ~3.3k | Automated safety check: Pass | MIT | |
| Observability Patternssoftspark/ai-toolkit | 179 | — | ~2.2k | Automated safety check: Pass | Apache-2.0 | |
| Telemetrymagnus919/agent-skills | 116 | — | ~3.9k | Automated safety check: Pass | MIT |
Similar skills
Observability Monitoring
AnastasiyaW/codex-claude-code-config
Design, audit, and troubleshoot production monitoring and observability using user-impact checks, layered telemetry, USE/RED, SLI/SLO/SLA, error budgets, cardinality controls, actionable alerting…
Observability Sre
majiayu000/spellbook
Observability and SRE expert. An agent skill from majiayu000/spellbook.
Observability Patterns
softspark/ai-toolkit
Observability: structured logs, metrics (RED/USE), tracing, SLO/SLI.
Telemetry
magnus919/agent-skills
Operate the observability stack that deploys as one unit: Prometheus scrape configuration, recording and alerting rules, relabeling, retention, and high availability; OpenTelemetry Collector…
Cloud Monitoring
seb1n/awesome-ai-agent-skills
Monitor cloud infrastructure and applications using metrics, logs, and traces to provide real-time observability into performance, health, and reliability.
More from prometheus/prometheus-mcp
Prometheus System Health Check
prometheus/prometheus-mcp
Builds a picture of whether Prometheus itself is healthy and successfully monitoring its targets, covering readiness, firing alerts, target health and TSDB load.
Prometheus Top Resource Consumers
prometheus/prometheus-mcp
Find the top CPU, memory, or disk consumers. Use for capacity reviews, noisy-neighbor hunts, and top-N questions about which jobs, pods, or instances use the…
Prometheus Cardinality Optimizer
prometheus/prometheus-mcp
Finds the metrics and labels behind a Prometheus series explosion using a connected Prometheus MCP server, then proposes relabeling, dropping or recording-rule fixes with measured impact.
Prometheus Rules Review
prometheus/prometheus-mcp
Audits Prometheus recording and alerting rules through a connected Prometheus MCP server, finds gaps and noisy alerts, and drafts improved rule-group YAML.
Prometheus Missing Data Troubleshooting
prometheus/prometheus-mcp
Finds where a Prometheus metric stops existing, whether at the target, the scrape, relabeling or the query, using the tools of a connected Prometheus MCP server.
Tune Prometheus Config
prometheus/prometheus-mcp
Review and tune Prometheus configuration and performance. An agent skill from prometheus/prometheus-mcp.
Related
Works with
Categories
Questions about Prometheus Error Rate Investigator
What does Prometheus Error Rate Investigator do?
Quantifies elevated error rates with PromQL, compares them to a baseline, and isolates which jobs or instances an error spike is concentrated in. The skill starts by discovering what error signals actually exist in the target system, since they vary: HTTP status counters, gRPC code counters, or dedicated errors or failures metrics, found with series or label-value queries and confirmed against metric metadata, alongside checking list_alerts and the ALERTS firing series for anything already flagged. From there it walks through computing a current error rate with rate() over raw counters, dividing by total traffic to get a ratio that separates more errors from more traffic, and running the same query over a longer range to find the inflection point against a known-good baseline window.
When should I use Prometheus Error Rate Investigator?
Prometheus Error Rate Investigator fits situations like: investigating a spike in 5xx or gRPC error responses; confirming whether a firing error-rate alert reflects a real traffic-wide problem; narrowing an error spike down to a specific job, instance or route.
How do I install Prometheus Error Rate Investigator in Claude Code?
Run `npx skills add prometheus/prometheus-mcp --skill investigate-error-rates -a claude-code`. Or copy the skill folder (pkg/mcp/assets/skills/investigate-error-rates in prometheus/prometheus-mcp) into .claude/skills/investigate-error-rates in your project. Claude Code loads it when a task matches its description.
How do I install Prometheus Error Rate Investigator in Codex?
Run `npx skills add prometheus/prometheus-mcp --skill investigate-error-rates -a codex`. Or copy the skill folder (pkg/mcp/assets/skills/investigate-error-rates in prometheus/prometheus-mcp) into .agents/skills/investigate-error-rates in your project. Codex loads it when a task matches its description.
Can I use Prometheus Error Rate Investigator in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add prometheus/prometheus-mcp --skill investigate-error-rates -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/investigate-error-rates, .gemini/skills/investigate-error-rates, .github/skills/investigate-error-rates and .opencode/skills/investigate-error-rates in your project.
What does Prometheus Error Rate Investigator need to run?
SKILL.md names no scripts, command-line tools or credentials: Prometheus Error Rate Investigator is instructions for the agent only. Our summary lists: A connected Prometheus MCP server. Compatibility (from SKILL.md): Requires the tools of a connected Prometheus MCP server.
Does Prometheus Error Rate Investigator access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Prometheus Error Rate Investigator safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Prometheus Error Rate Investigator use?
Prometheus Error Rate Investigator is published under the Apache-2.0 licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Prometheus Error Rate Investigator use?
About 592 tokens (SKILL.md is roughly 2.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Prometheus Error Rate Investigator?
Skills that share tags, products or a category with Prometheus Error Rate Investigator: Monitoring Observability (ahmedasmar/devops-claude-skills, 203 stars), Observability Monitoring (AnastasiyaW/codex-claude-code-config, 154 stars), Observability Sre (majiayu000/spellbook, 287 stars) and Observability Patterns (softspark/ai-toolkit, 179 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Prometheus Error Rate Investigator?
prometheus (a GitHub organization) maintains it in prometheus/prometheus-mcp, which has 120 GitHub stars. The repository holds 7 skills in this directory. The repository was last updated on October 4, 2026.
Source: prometheus/prometheus-mcp on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.