New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Chaos Experiment

by harness in harness/harness-skills

A skill your agent uses when the user asks to create, edit, update, design, or configure a Harness Chaos Experiment — including faults, probes, actions, experiment YAML, fault injection, pod-delete…

Apache-2.0Auto-check passedDevOps & Cloud

Install Chaos Experiment

skills CLI
$ npx skills add harness/harness-skills --skill chaos-experiment -a claude-code

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

GitHub CLI
$ gh skill install harness/harness-skills chaos-experiment --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/harness/harness-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/chaos-experiment .claude/skills/chaos-experiment && 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
chaos-experiment
GitHub stars
115
Token cost
~1.6k tokens
SKILL.md length
716 words
Files
4 (incl. references)
Skills in repo
24
Repo updated
First seen
Licence
Apache-2.0

At a glance

A skill your agent uses when the user asks to create, edit, update, design, or configure a Harness Chaos Experiment — including faults, probes, actions, experiment YAML, fault injection, pod-delete…

  • Works in 2 steps: Establish scope → Route before acting
  • The user asks to create
  • SKILL.md covers Instructions, Examples, Performance Notes and Troubleshooting
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Chaos Experiment is an agent skill from harness/harness-skills. Use when the user asks to create, edit, update, design, or configure a Harness Chaos Experiment — including faults, probes, actions, experiment YAML, fault injection, pod-delete, or resilience tests. Do not use for Chaos steps inside a pipeline or a DRTest stage; use chaos-dr-test for those. Do not use to run an experiment, list past runs, or inspect run results — call the Chaos MCP tools directly for those (see Performance Notes). Trigger phrases: chaos experiment, chaos engineering, resilience test, fault…

Its SKILL.md is about 1.6k tokens, which your agent loads only when the skill is triggered. The skill folder holds 4 other files, including reference files (for example `references/components.md`, `references/create.md` and `references/edit.md`). Compatibility notes: Requires Harness MCP v2 server (harness-mcp-v2)

It sits in DevOps & Cloud, covering Chaos engineering and MCP servers. The repository describes itself as: A collection of structured AI agent skills that enable Claude Code, Cursor, GitHub Copilot, and other AI coding assistants to create, operate, debug, and govern Harness CI/CD… The licence is Apache-2.0.

When your agent uses it

  • The user asks to create
  • Configure a Harness Chaos Experiment — including faults
  • Experiment YAML
  • Fault injection

Example prompts

  • “/chaos-experiment”

Requirements

  • Compatibility (from SKILL.md): Requires Harness MCP v2 server (harness-mcp-v2)

Workflow steps

2 steps, taken from the step headings in SKILL.md.

  1. Establish scope
  2. Route before acting

What it can do on your machine

Read from SKILL.md and the folder at commit c25faee. 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 Harness MCP v2 server (harness-mcp-v2)

    From compatibility in the SKILL.md frontmatter.

Context cost

Chaos Experiment loads about 1.6k tokens when it runs, and up to ~19k if it reads all its reference files. Until then it costs about 143 tokens; SKILL.md has 716 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~143
When it runs · the whole SKILL.md, loaded when a task matches
~1.6k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~19k

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 harness/harness-skills at commit c25faee, republished under its Apache-2.0 licence (© harness). 716 words, ~1,609 tokens.

Download SKILL.mdSave it as .claude/skills/chaos-experiment/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.
name
chaos-experiment
description
Use when the user asks to create, edit, update, design, or configure a Harness Chaos Experiment — including faults, probes, actions, experiment YAML, fault injection, pod-delete, or resilience tests. Do not use for Chaos steps inside a pipeline or a DRTest stage; use chaos-dr-test for those. Do not use to run an experiment, list past runs, or inspect run results — call the Chaos MCP tools directly for those (see Performance Notes). Trigger phrases: chaos experiment, chaos engineering, resilience test, fault injection, pod-delete, chaos hub template.
compatibility
Requires Harness MCP v2 server (harness-mcp-v2)
metadata.author
Harness
metadata.version
3.0.0
metadata.mcp-server
harness-mcp-v2
metadata.module
CHAOS
license
Apache-2.0

Chaos Experiment

Create, edit, and answer reference questions about Harness Chaos Experiments — faults, probes, actions, and the underlying ChaosExperiment manifest — via MCP.

MANDATORY — Tool Result Verification:

  • NEVER claim a tool call succeeded without an explicit success response from the tool.
  • NEVER fabricate, assume, or pre-empt tool results. If you called a tool and did not receive a confirmed result, say so and retry.
  • When a tool returns a review/elicitation prompt (e.g., "Waiting for user to review before update..."), the user's approval does NOT mean the operation completed. You MUST wait for the follow-up tool response confirming execution before reporting success.
  • If the tool response is ambiguous or missing, tell the user honestly and offer to retry.

Instructions

Step 1: Establish scope

Before any harness_list, harness_get, harness_create, harness_update, or harness_execute call in this skill, establish org/project scope using the repo-wide playbook at references/scope-establishment.md:

  • Parse org / project from the user's message or a pasted Harness UI URL first.
  • Ask only for what is missing — do not re-ask for values already known.
  • Restate the active scope (e.g. org=default, project=payments) before the first mutating call.
  • Never assume a default org/project unless the user explicitly confirms it.
Step 2: Route before acting

Choose the first matching entry route below. Read that entry file first; it may direct you to read additional shared references later.

  • Reference question — the user asks how a fault, probe, action, target workload, tunable, vertex, manifest field, or YAML structure works, without asking to create or modify an experiment: Read references/components.md. No scaffolded YAML is required for a reference-only answer.
  • Existing experiment — the user explicitly asks to edit, update, modify, change, rename, fix, remove from, or add to an existing experiment, says the experiment already exists, or provides an existing experiment UUID: Read references/edit.md. Preserve experimentId and identity. Do not generate a new UUID. Saving an edit reuses harness_create as an upsert — chaos_experiment has no separate update operation; do not call harness_update for this resource type.
  • New experiment — the user asks to create, build, set up, make, design, or launch a new experiment: Read references/create.md. This route still applies when the user supplies the desired name for the new experiment. Generate a new UUID; do not load an existing experiment.
  • Unsure — ask: "Would you like to create a new chaos experiment, edit an existing one, or ask a reference question about experiment YAML/components?"

Fault / probe / action add-remove mechanics live in references/components.md. Create and edit flows read it after working YAML exists; reference-only questions may read it directly.

For ChaosFault / ChaosProbe / ChaosAction / Chaos steps in a pipeline or DRTest stage, stop and use chaos-dr-test instead.

Show full SKILL.md (284 more words)Show less

Examples

  • "Create a pod-delete chaos experiment on my checkout-service" -> route to references/create.md
  • "Add an HTTP probe to my test-exp-007 experiment" -> route to references/edit.md
  • "Edit the experiment with UUID 218b1053-..., remove the CPU hog fault" -> route to references/edit.md
  • "What does the vertices array in a ChaosExperiment manifest mean?" -> route to references/components.md (reference-only, no YAML changes)
  • "How do I target a workload by labels instead of name?" -> route to references/components.md

Performance Notes

  • Do not guess fault, probe, or action identities, infrastructure IDs, or runtime variable names. Always call the matching harness_list / harness_get / harness_execute(action="get_variables") and use the returned values.
  • Validate the manifest against the YAML Validation Rules in references/components.md before presenting it or saving — no duplicate keys, no orphan vertices references, no blank lines.
  • Wait for the tool's confirmed save response before reporting success; a review/elicitation prompt is not a completed save.
  • Not covered by this skill: running an experiment, listing existing experiments, or inspecting run results. Use the Chaos MCP tools directly — harness_execute(resource_type="chaos_experiment", action="run", resource_id="<experiment_id>", ...), harness_list(resource_type="chaos_experiment_run", ...), harness_get(resource_type="chaos_experiment_run", resource_id="<run_id>", ...).

Troubleshooting

  • Experiment fails to save / validation error — call harness_describe(resource_type="chaos_experiment") to check the authoritative create body schema, fix the offending field, and retry.
  • No enterprise or custom faults returned — confirm the selected infrastructure type (Kubernetes -> KubernetesV2, Linux, Windows) matches the fault's supported infrastructure; re-run harness_list(resource_type="chaos_fault", ...) with the corrected infrastructure filter.
  • Target workload discovery returns empty — the chaos infrastructure's Service Discovery agent may not have scanned the namespace yet; ask the user to verify the infrastructure is ACTIVE and isChaosEnabled: true before retrying discovered_service / discovered_namespace calls.
  • Ambiguous create vs. edit intent — if the user's request could mean either, ask explicitly per Step 2 above rather than guessing; creating a duplicate experiment is harder to undo than asking one clarifying question.

© harness, 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

SKILL.md and 3 other files (references) in skills/chaos-experiment of harness/harness-skills.

  • SKILL.md
  • references/components.md
  • references/create.md
  • references/edit.md

Open the folder on GitHubat commit c25faee

Compare with similar skills

Chaos Experiment 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.

Chaos Experiment compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Chaos Experiment this skillharness/harness-skills115—~1.6kAutomated safety check: PassApache-2.0
Agent Harness Fault Injectionsickn33/agentic-awesome-skills47k1 repos~2.9kAutomated safety check: PassMIT
AWS Cdk Developmentzxkane/aws-skills3672 repos~2.5kAutomated safety check: PassMIT
Kubernetes Network Root Cause Analysiskubeshark/kubeshark12k—~5.3kAutomated safety check: PassApache-2.0
Executing Distributed System Testsshenli/distributed-system-testing231—~5.1kAutomated safety check: NotesMIT
K8s Agent Sandbox MCPkubernetes-sigs/agent-sandbox4.2k—~1.3kAutomated safety check: PassApache-2.0

Similar skills

  • Agent Harness Fault Injection

    sickn33/agentic-awesome-skills

    A skill your agent uses when an agent workflow needs deterministic recovery evidence for sandbox, MCP/tool, worker, checkpoint, memory, or orchestration failures.

    47k GitHub starsUsed in 1 repo~2.9k tokens
    Agent WorkflowsAuto-check passed

  • AWS Cdk Development

    zxkane/aws-skills

    AWS Cloud Development Kit (CDK) expert for building cloud infrastructure with TypeScript/Python.

    367 GitHub starsUsed in 2 repos~2.5k tokens
    DevOps & CloudAuto-check passed

  • Kubernetes Network Root Cause Analysis

    kubeshark/kubeshark

    Investigates past Kubernetes incidents from Kubeshark traffic snapshots: takes captures, dissects API calls, extracts PCAPs and compares traffic over time.

    12k GitHub stars~5.3k tokensUpdated 2 days ago
    DevOps & CloudAuto-check passed

  • Executing Distributed System Tests

    shenli/distributed-system-testing

    A skill your agent uses when running a previously designed distributed-systems test plan against a real or simulated cluster — driving fault injection, workload, chaos scenarios, linearizability /…

    231 GitHub stars~5.1k tokensUpdated 2 mo ago
    DevOps & CloudAuto-check: notes

  • K8s Agent Sandbox MCP

    kubernetes-sigs/agent-sandbox

    Official

    An MCP server skill for managing Kubernetes sandboxes. An agent skill from kubernetes-sigs/agent-sandbox.

    4.2k GitHub stars~1.3k tokensUpdated yesterday
    DevOps & CloudAuto-check passed

  • Greptimedb Perses Dashboard

    GreptimeTeam/dashboard

    Generate Perses dashboards or single panels for GreptimeDB. An agent skill from GreptimeTeam/dashboard.

    111 GitHub stars~3.9k tokensUpdated today
    DevOps & CloudAuto-check passed

More from harness/harness-skills

All 24 skills in this repo

  • Audit Report

    harness/harness-skills

    Generate audit reports and compliance trails using Harness audit trail data via MCP v2 tools.

    115 GitHub stars~1.3k tokensUpdated 3 days ago
    Auto-check passed

  • Chaos Dr Test

    harness/harness-skills

    A skill your agent uses when working with Chaos Engineering steps inside a Harness pipeline.

    115 GitHub stars~2.6k tokensUpdated 3 days ago
    Auto-check passed

  • Cleanup Feature Flags

    harness/harness-skills

    Remove a launched Harness FME feature flag from application code, keeping the treatment FME serves today, and open a pull request.

    115 GitHub stars~2.4k tokensUpdated 3 days ago
    Auto-check passed

  • Configure Repo Scan

    harness/harness-skills

    Configure code scanning in Harness pipelines using STO security scanners.

    115 GitHub stars~2.2k tokensUpdated 3 days ago
    Auto-check passed

  • Create Agent Template

    harness/harness-skills

    Generate Harness Agent Template files for AI-powered automation agents.

    115 GitHub stars~2.2k tokensUpdated 3 days ago
    Auto-check passed

  • Create Metric

    harness/harness-skills

    Create a Harness FME metric: helps decide what to measure, suggests candidate metrics from application code, resolves traffic type and event type IDs, drafts the payload, then creates.

    115 GitHub stars~3.4k tokensUpdated 3 days ago
    Auto-check passed

Categories

Questions about Chaos Experiment

What does Chaos Experiment do?

A skill your agent uses when the user asks to create, edit, update, design, or configure a Harness Chaos Experiment — including faults, probes, actions, experiment YAML, fault injection, pod-delete…. Chaos Experiment is an agent skill from harness/harness-skills. Use when the user asks to create, edit, update, design, or configure a Harness Chaos Experiment — including faults, probes, actions, experiment YAML, fault injection, pod-delete, or resilience tests.

When should I use Chaos Experiment?

Chaos Experiment fits situations like: the user asks to create; configure a Harness Chaos Experiment — including faults; experiment YAML; fault injection.

How do I install Chaos Experiment in Claude Code?

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

How do I install Chaos Experiment in Codex?

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

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

What does Chaos Experiment need to run?

SKILL.md names no scripts, command-line tools or credentials: Chaos Experiment is instructions for the agent only. Compatibility (from SKILL.md): Requires Harness MCP v2 server (harness-mcp-v2).

Does Chaos Experiment 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 Chaos Experiment 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 Chaos Experiment use?

Chaos Experiment 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 Chaos Experiment use?

About 1.6k tokens (SKILL.md is roughly 6.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 17k tokens, read only when the agent opens those files.

What are the alternatives to Chaos Experiment?

Skills that share tags, products or a category with Chaos Experiment: Agent Harness Fault Injection (sickn33/agentic-awesome-skills, 47k stars), AWS Cdk Development (zxkane/aws-skills, 367 stars), Kubernetes Network Root Cause Analysis (kubeshark/kubeshark, 12k stars) and Executing Distributed System Tests (shenli/distributed-system-testing, 231 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Chaos Experiment?

harness (a GitHub organization) maintains it in harness/harness-skills, which has 115 GitHub stars. The repository holds 24 skills in this directory. The repository was last updated on October 6, 2026.

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