Agent skill

Observability

by WrongStack in WrongStack/WrongStack

A skill your agent uses when adding or reviewing logging, metrics, or tracing in an application, or when a production problem can't be diagnosed from the signals that exist today.

MITAuto-check passedDevOps & Cloud

Install Observability

skills CLI
$ npx skills add WrongStack/WrongStack --skill observability -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install WrongStack/WrongStack observability --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/WrongStack/WrongStack.git skills-src && mkdir -p .claude/skills && cp -r skills-src/packages/core/skills/observability .claude/skills/observability && rm -rf skills-src

Use ~/.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/

Facts

Skill name
observability
GitHub stars
370
Token cost
~1.2k tokens
SKILL.md length
450 words
Files
2
Skills in repo
38
Repo updated
First seen
Licence
MIT

At a glance

A skill your agent uses when adding or reviewing logging, metrics, or tracing in an application, or when a production problem can't be diagnosed from the signals that exist today.

  • Works in 8 steps: Use the project's existing logger and… → Log structured events, not sentences: a… → Levels mean something: error needs a… → …
  • Reviewing logging
  • SKILL.md covers Overview, Rules, What to instrument and Patterns, plus 3 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Observability is an agent skill from WrongStack/WrongStack. Use this skill when adding or reviewing logging, metrics, or tracing in an application, or when a production problem can't be diagnosed from the signals that exist today. Triggers: user says "logging", "logs", "trace", "tracing", "metrics", "observability", "instrument", "OpenTelemetry", "structured logging", "log level", "correlation id", "monitoring", "alert".

Its SKILL.md is about 1.2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 1 other file (for example `SKILL.save.md`).

It sits in DevOps & Cloud, covering Observability. It works with OpenTelemetry. The repository describes itself as: An AI coding agent that reads your code, edits files, runs commands, and reasons through bugs — across a terminal REPL, a full-screen TUI, and a browser UI, while you keep your… The licence is MIT.

When your agent uses it

  • Reviewing logging
  • Tracing in an application
  • A production problem cant be diagnosed from the signals that exist today

Example prompts

  • “t be diagnosed from the signals that exist today. Triggers: user says”
  • “tracing”
  • “metrics”
  • “/observability”

Requirements

  • Node.js

Workflow steps

8 steps, taken from the first numbered list in SKILL.md.

  1. Use the project's existing logger and conventions; find how a neighbouring
  2. Log structured events, not sentences: a stable event name plus fields
  3. Levels mean something: error needs a human, warn is degraded but
  4. Correlate: carry a request or trace id through async boundaries
  5. Never log secrets, tokens, credentials, or personal data. Configure redaction
  6. Log an error once, where it is handled, with the error object and context —
  7. Metrics use bounded label values. User ids, raw URLs, and error messages as
  8. Trace the I/O: spans around outbound HTTP, database, queue, and cache calls,

What it can do on your machine

Read from SKILL.md and the folder at commit 57f6018. 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 (its code samples are typescript).

    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.

Context cost

Observability loads about 1.2k tokens when it runs. Until then it costs about 95 tokens; SKILL.md has 450 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~95
When it runs · the whole SKILL.md, loaded when a task matches
~1.2k

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 WrongStack/WrongStack at commit 57f6018, republished under its MIT licence (© WrongStack). 450 words, ~1,222 tokens.

Download SKILL.mdSave it as .claude/skills/observability/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
observability
description
Use this skill when adding or reviewing logging, metrics, or tracing in an application, or when a production problem can't be diagnosed from the signals that exist today. Triggers: user says "logging", "logs", "trace", "tracing", "metrics", "observability", "instrument", "OpenTelemetry", "structured logging", "log level", "correlation id", "monitoring", "alert".
version
2.0.0
required-capabilities
filesystem.read, filesystem.write
optional-capabilities
code.inspect

Observability

Overview

Instrument so that the next incident can be answered from telemetry instead of by adding logs and redeploying. Work with what the project already has — its logger (pino, winston, structlog, slog, log/zap), metrics client, and any OpenTelemetry setup. Adding a second logging stack is almost always wrong.

Rules

  1. Use the project's existing logger and conventions; find how a neighbouring module logs before adding a line.
  2. Log structured events, not sentences: a stable event name plus fields (logger.info({ orderId, durationMs }, 'order.charged')), so logs can be filtered and aggregated.
  3. Levels mean something: error needs a human, warn is degraded but handled, info is a meaningful business or lifecycle event, debug is off in production.
  4. Correlate: carry a request or trace id through async boundaries (OpenTelemetry context or AsyncLocalStorage) and attach it to every log line.
  5. Never log secrets, tokens, credentials, or personal data. Configure redaction once in the logger (for example pino redact paths), not per call site.
  6. Log an error once, where it is handled, with the error object and context — not at every layer it passes through.
  7. Metrics use bounded label values. User ids, raw URLs, and error messages as labels explode cardinality; use route templates and error classes.
  8. Trace the I/O: spans around outbound HTTP, database, queue, and cache calls, with status recorded on failure. Prefer auto-instrumentation where it exists.
Show full SKILL.md (219 more words)Show less

What to instrument

SignalForExamples
LogsWhat happened to one request or jobpayment.failed with order id, provider code, attempt
MetricsHow the system behaves in aggregateRequest rate, error rate, latency histogram per route (RED); queue depth, pool saturation (USE)
TracesWhere the time went across callsSpan per inbound request and per outbound dependency

Start from the question an on-call engineer will ask — "why did checkout fail for this customer?", "which dependency made p99 spike?" — and make sure the answer is recorded.

Patterns

ts
// Redaction configured once, at the logger.
const logger = pino({
  level: process.env.LOG_LEVEL ?? 'info',
  redact: ['req.headers.authorization', 'req.headers.cookie', '*.password', '*.token'],
});

// One structured event, with correlation and the error object.
try {
  await chargeCard(order);
  logger.info({ orderId: order.id, durationMs: Date.now() - started }, 'order.charged');
} catch (err) {
  logger.error({ err, orderId: order.id, traceId: currentTraceId() }, 'order.charge_failed');
  throw new PaymentError('charge failed', { cause: err });
}
ts
// A span around an outbound dependency.
const tracer = trace.getTracer('checkout');

export async function reserveStock(sku: string, qty: number): Promise<void> {
  await tracer.startActiveSpan('inventory.reserve', async (span) => {
    span.setAttributes({ 'inventory.sku': sku, 'inventory.qty': qty });
    try {
      await inventoryClient.reserve(sku, qty);
    } catch (err) {
      span.recordException(err as Error);
      span.setStatus({ code: SpanStatusCode.ERROR });
      throw err;
    } finally {
      span.end();
    }
  });
}

Anti-patterns

  • console.log in a codebase with a logger — unstructured, unleveled, unredacted.
  • Logging and rethrowing at every layer — one failure becomes ten error lines.
  • Logging whole request or user objects — the fastest path to leaking personal data.
  • High-cardinality metric labels — breaks the metrics backend and the bill.
  • Alerts on causes instead of symptoms — page on user-facing error rate and latency, not on CPU.

Before returning

  • Uses the project's existing logger, metrics, and tracing setup
  • Structured events with stable names; levels used deliberately
  • Correlation id present across async boundaries
  • No secrets or personal data; redaction configured centrally
  • Errors logged once, at the handling site
  • Metric labels bounded; outbound I/O traced

Skills in scope

  • security-scanner — for confirming nothing sensitive reaches logs
  • data-governance — for retention and personal-data classification of telemetry
  • node-modern — for async context propagation in Node.js

© WrongStack, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in packages/core/skills/observability of WrongStack/WrongStack.

  • SKILL.md
  • SKILL.save.md

Open the folder on GitHubat commit 57f6018

Compare with similar skills

Observability 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.

Observability compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Observability this skillWrongStack/WrongStack370—~1.2kAutomated safety check: PassMIT
Motel Debugkitlangton/motel298—~2.2kAutomated safety check: PassMIT
Tempsgotempsh/temps826—~1.9kAutomated safety check: PassApache-2.0
Axiom Metrics Queryopenclaw/clawhub9.5k—~2.6kAutomated safety check: PassMIT
UModel Root Cause Analysisalibaba/UnifiedModel412—~1.9kAutomated safety check: PassCustom licence
Agent Kill Switchvivekchand/clawmetry425—~1.1kAutomated safety check: PassMIT

Similar skills

  • Motel Debug

    kitlangton/motel

    Debug applications with motel, a local OpenTelemetry ingest and query server.

    298 GitHub stars~2.2k tokensUpdated 1 mo ago
    DevOps & CloudAuto-check passed
  • Temps

    gotempsh/temps

    Manage, deploy, operate, and instrument applications with Temps.

    826 GitHub stars~1.9k tokensUpdated today
    DevOps & CloudAuto-check passed
  • Axiom Metrics Query

    openclaw/clawhub

    Explores and queries OpenTelemetry metrics in Axiom MetricsDB, listing datasets, metrics and tags first and picking the right aggregation for each metric's type.

    9.5k GitHub stars~2.6k tokensUpdated yesterday
    DevOps & CloudAuto-check passed
  • UModel Root Cause Analysis

    alibaba/UnifiedModel

    Investigates a service incident to its root cause by querying a UModel object graph alongside metrics, logs, topology and recent deployments.

    412 GitHub stars~1.9k tokensUpdated 14 days ago
    DevOps & CloudAuto-check passed
  • Agent Kill Switch

    vivekchand/clawmetry

    Give the human an off switch and a cost meter for the coding agents on this machine, using ClawMetry.

    425 GitHub stars~1.1k tokensUpdated today
    DevOps & CloudAuto-check passed
  • Agentmeasure

    roy-tong/AgentMeasure

    Check whether agent telemetry preserves measurement semantics.

    218 GitHub stars~753 tokensUpdated today
    DevOps & CloudAuto-check passed

More from WrongStack/WrongStack

All 38 skills in this repo
  • Design Craft

    WrongStack/WrongStack

    Design or substantially improve user-facing interfaces with a product-specific visual direction, content hierarchy, and rendered critique.

    370 GitHub stars~2k tokensUpdated yesterday
    Auto-check passed
  • Design Critique

    WrongStack/WrongStack

    A skill your agent uses to audit an interface that already exists and say precisely why it looks generated, templated, or unfinished — a scored rubric across composition, typography, color, states…

    370 GitHub stars~3k tokensUpdated yesterday
    Auto-check passed
  • Mailbox Bridge

    WrongStack/WrongStack

    A skill your agent uses when external coding agents (Claude Code, Aider, custom scripts) need to participate in the project's shared WrongStack mailbox, or when a user asks to "expose the mailbox"…

    370 GitHub stars~3.9k tokensUpdated yesterday
    Auto-check passed
  • Multi Agent

    WrongStack/WrongStack

    A skill your agent uses whenever work can be split across multiple AI agents running in parallel, or when orchestrating leader/worker patterns in WrongStack.

    370 GitHub stars~3.6k tokensUpdated yesterday
    Auto-check passed
  • Web Platform Baseline

    WrongStack/WrongStack

    Use this skill before asserting that a CSS, HTML or accessibility capability is available, unavailable, or the right tool — it carries dated, refreshable platform facts and refuses to let stale…

    370 GitHub stars~1.6k tokensUpdated yesterday
    Auto-check passed
  • Wrongstack Mailbox

    WrongStack/WrongStack

    A skill your agent uses when the user wants to communicate with WrongStack's shared project mailbox from outside WrongStack — read messages sent by WrongStack agents, send replies, broadcast to all…

    370 GitHub stars~3.5k tokensUpdated yesterday
    Auto-check passed

Works with

Categories

Questions about Observability

What does Observability do?

A skill your agent uses when adding or reviewing logging, metrics, or tracing in an application, or when a production problem can't be diagnosed from the signals that exist today. Observability is an agent skill from WrongStack/WrongStack. Use this skill when adding or reviewing logging, metrics, or tracing in an application, or when a production problem can't be diagnosed from the signals that exist today.

When should I use Observability?

Observability fits situations like: reviewing logging; tracing in an application; A production problem cant be diagnosed from the signals that exist today.

How do I install Observability in Claude Code?

Run `npx skills add WrongStack/WrongStack --skill observability -a claude-code`. Or copy the skill folder (packages/core/skills/observability in WrongStack/WrongStack) into .claude/skills/observability in your project. Claude Code loads it when a task matches its description.

How do I install Observability in Codex?

Run `npx skills add WrongStack/WrongStack --skill observability -a codex`. Or copy the skill folder (packages/core/skills/observability in WrongStack/WrongStack) into .agents/skills/observability in your project. Codex loads it when a task matches its description.

Can I use Observability 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 WrongStack/WrongStack --skill observability -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/observability, .gemini/skills/observability, .github/skills/observability and .opencode/skills/observability in your project.

What does Observability need to run?

SKILL.md names no scripts, command-line tools or credentials: Observability is instructions for the agent only. Our summary lists: Node.js.

Does Observability 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 Observability 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 Observability use?

Observability is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Observability use?

About 1.2k tokens (SKILL.md is roughly 4.9k 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 Observability?

Skills that share tags, products or a category with Observability: Motel Debug (kitlangton/motel, 298 stars), Temps (gotempsh/temps, 826 stars), Axiom Metrics Query (openclaw/clawhub, 9.5k stars) and UModel Root Cause Analysis (alibaba/UnifiedModel, 412 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Observability?

WrongStack (a GitHub organization) maintains it in WrongStack/WrongStack, which has 370 GitHub stars. The repository holds 38 skills in this directory. The repository was last updated on October 7, 2026.

Source: WrongStack/WrongStack on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.