agentleFS
Sign inStart free
thisrohanguptathisrohangupta/harness-skills/skills/chaos-experiment/SKILL.md

chaos-experiment

Skill2 starsChanged 7 months agoApache-2.0Archived repositoryView on GitHub

Create and manage chaos experiments using Harness Chaos Engineering via MCP. Run resilience tests like pod deletion, CPU stress, and network faults. Use when user says "chaos experiment", "chaos engineering", "resilience test", "chaos test", or wants to test system reliability.

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 11 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈11

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create and manage chaos experiments using Harness Chaos Engineering via MCP.

    line 14
  2. Distinctive:

    Pod Network Latency: - Add artificial latency

    line 106
  3. Distinctive:

    EC2 Stop: - Stop AWS EC2 instances

    line 108
  4. Distinctive:

    ECS Task Stop: - Stop ECS tasks

    line 109
  5. Distinctive:

    Wait for experiment completion before analyzing results.

    line 131
  6. Distinctive:

    Verify the target infrastructure and service are healthy before running chaos experiments.

    line 132
  7. Distinctive:

    Verify chaos infrastructure is connected and active

    line 137
  8. Distinctive:

    Check target application/namespace exists

    line 138
  9. Distinctive:

    Ensure RBAC permissions for chaos operations

    line 139
  10. Distinctive:

    Check probe endpoints are accessible

    line 142
  11. Distinctive:

    Verify probe timeout settings

    line 143

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Chaos Experiment

Create and manage chaos experiments using Harness Chaos Engineering via MCP.

Instructions

Step 1: Check Infrastructure

Call MCP tool: harness_list
Parameters:
  resource_type: "chaos_infrastructure"
  org_id: "<organization>"
  project_id: "<project>"

Step 2: Browse Templates

Call MCP tool: harness_list
Parameters:
  resource_type: "chaos_experiment_template"
  org_id: "<organization>"
  project_id: "<project>"

Step 3: List Existing Experiments

Call MCP tool: harness_list
Parameters:
  resource_type: "chaos_experiment"
  org_id: "<organization>"
  project_id: "<project>"

Step 4: Create Experiment

Call MCP tool: harness_create
Parameters:
  resource_type: "chaos_experiment"
  org_id: "<organization>"
  project_id: "<project>"
  body: <experiment definition>

Step 5: Run Experiment

Call MCP tool: harness_execute
Parameters:
  resource_type: "chaos_experiment"
  action: "run"
  resource_id: "<experiment_id>"
  org_id: "<organization>"
  project_id: "<project>"

Step 6: Monitor Results

Call MCP tool: harness_list
Parameters:
  resource_type: "chaos_experiment_run"
  org_id: "<organization>"
  project_id: "<project>"

Get specific run details:

Call MCP tool: harness_get
Parameters:
  resource_type: "chaos_experiment_run"
  resource_id: "<run_id>"

Step 7: Check Probes

Call MCP tool: harness_list
Parameters:
  resource_type: "chaos_probe"
  org_id: "<organization>"
  project_id: "<project>"

Common Experiment Types

  • Pod Delete - Kill pods to test recovery
  • Pod CPU Hog - Stress CPU to test throttling
  • Pod Memory Hog - Consume memory to test OOM handling
  • Pod Network Loss - Simulate network failures
  • Pod Network Latency - Add artificial latency
  • Node Drain - Drain K8s nodes
  • EC2 Stop - Stop AWS EC2 instances
  • ECS Task Stop - Stop ECS tasks

Chaos Resource Types

Resource TypeOperationsDescription
chaos_experimentlist, get, create, update, delete, runExperiments
chaos_experiment_runlist, getRun history/results
chaos_experiment_templatelist, getPre-built templates
chaos_infrastructurelist, getTarget infrastructure
chaos_probelist, getHealth probes

Examples

  • "Show me all chaos experiments" - List chaos_experiment
  • "Create a pod-delete experiment for checkout-service" - Create chaos_experiment
  • "Run the weekly resilience test" - Execute run action
  • "What were the results of the last chaos run?" - Get chaos_experiment_run

Performance Notes

  • Review existing experiments before creating duplicates. Check for similar fault types targeting the same service.
  • Wait for experiment completion before analyzing results. Do not draw conclusions from partial runs.
  • Verify the target infrastructure and service are healthy before running chaos experiments.

Troubleshooting

Experiment Won't Run

  • Verify chaos infrastructure is connected and active
  • Check target application/namespace exists
  • Ensure RBAC permissions for chaos operations

Probes Failing

  • Check probe endpoints are accessible
  • Verify probe timeout settings
  • Review probe type matches expected behavior

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free