Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Network Simulation Skill · a5c-ai | DecimalAI

a5c-ai/network-simulation

SkillScore100Maintained

Skill for network condition simulation, emulation, and chaos engineering

DevOpsby a5c-ai

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

DevOps & SRE pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/a5c-ai-network-simulation/SKILL.md

decimalai skills pull a5c-ai-network-simulation --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/a5c-ai-network-simulation/SKILL.md https://app.decimal.ai/s/a5c-ai-network-simulation/SKILL.md

Raw file: /s/a5c-ai-network-simulation/SKILL.md ↗ · machine-readable summary at /s/a5c-ai-network-simulation.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

23 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 25d ago

Yes — +17% lift+17% tokens0% turns100% of whole cases pass with skill (23/23)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-01	✗→✓	▲ Improved	68%	0%
case-08	✗→✓	▲ Improved	-57%	0%
case-16	✗→✓	▲ Improved	-50%	0%
case-17	✗→✓	▲ Improved	40%	0%
case-02	✓→✓	＝ Same ✓	77%	0%

Showing 5 of 23 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/10/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

a5c-ai / babysitter

library/specializations/network-programming/skills/network-simulation/SKILL.md

View on GitHub ↗

License MITImported 8/10/2026Version v1

Skill contents

SKILL.md

Network Simulation Skill

Expert skill for network condition simulation, emulation, and chaos engineering testing environments.

Capabilities

Traffic Control: Configure tc (traffic control) for latency, bandwidth, and packet loss simulation
Network Namespaces: Set up network namespaces for network isolation testing
WAN Emulation: Emulate WAN conditions with netem (delay, jitter, loss, corruption)
Virtual Topologies: Create virtual network topologies for testing
Packet Manipulation: Simulate packet loss, reordering, duplication, and corruption
Degraded Conditions Testing: Test applications under degraded network conditions
Chaos Scenarios: Generate chaos engineering scenarios for resilience testing
Container Networking: Configure Docker/Kubernetes network simulation

Tools and Dependencies

tc - Traffic control for Linux
netem - Network emulator
ip netns - Network namespaces
mininet - Network emulator
toxiproxy - TCP proxy for chaos testing
comcast - Network simulation tool
pumba - Container chaos testing

Target Processes

network-testing-framework.js
load-testing-tool.js
protocol-fuzzer.js
tcp-socket-server.js

Usage Examples

Latency Simulation

bash
tc qdisc add dev eth0 root netem delay 100ms 20ms distribution normal
tc qdisc change dev eth0 root netem delay 200ms 40ms 25% correlation

Packet Loss Simulation

bash
tc qdisc add dev eth0 root netem loss 5% 25%
tc qdisc add dev eth0 root netem loss gemodel 1% 10% 70% 0.1%

Bandwidth Limiting

bash
tc qdisc add dev eth0 root tbf rate 1mbit burst 32kbit latency 400ms

Network Namespace Isolation

bash
ip netns add test_ns
ip link add veth0 type veth peer name veth1
ip link set veth1 netns test_ns
ip netns exec test_ns ip addr add 10.0.0.2/24 dev veth1
ip netns exec test_ns ip link set veth1 up

Toxiproxy Chaos

bash
toxiproxy-cli create -l localhost:6379 -u localhost:6380 redis_proxy
toxiproxy-cli toxic add -t latency -a latency=1000 redis_proxy

Quality Gates

Network conditions applied correctly
Application behavior under degraded conditions verified
Chaos scenarios executed successfully
Recovery testing completed
Performance baselines established

Illustrative Want to see what a5c-ai/network-simulation would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

8/30/2026, 08:58 UTC · 23 cases

Lift

+17%

100% vs 83%

Agent turns

0%

1.0 vs 1.0

Tokens

+17%

1,937.55 vs 1,656.32

fail → pass · 4pass → pass · 19

Per-case results (23)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-01 We want to simulate WAN latency on Linux network interface eth0 with a baseline of 100ms delay and a jitter of 20ms. A developer suggested using a uniform distribution default or Pareto distribution, but we specifically require a Gaussian distribution curve for realistic WAN variance. Write the exact traffic control command to add a 100ms delay with 20ms variance using a normal distribution.	fail→pass	10,967	9,060	-17%	1	1	0%	854	1,435	+68%	0	0	—
case-02 We have an existing root netem qdisc on eth0 and need to update it to increase latency to 200ms with 40ms jitter, while making delay values correlated with the previous packet delay by 25%. A team member tried to use 'tc qdisc add' which failed because the qdisc already exists. Provide the tc command to modify the existing netem latency settings with correlation.	pass→pass	19,834	9,355	-53%	1	1	0%	812	1,441	+77%	0	0	—
case-03 We need to simulate simple random packet loss on network interface eth0 where 5% of packets are dropped on average, with a 25% correlation to previous drop events. Write the tc netem command to apply this loss pattern.	pass→pass	8,709	8,363	-4%	1	1	0%	709	1,206	+70%	0	0	—
case-04 To test application resilience against bursty network drops, we want to simulate packet loss using a Gilbert-Elliot model instead of uniform random loss. The state parameters required are p=1%, r=10%, 1-h=70%, and 1-k=0.1%. Write the tc netem command for eth0 using this specific state-based loss model.	pass→pass	11,718	9,707	-17%	1	1	0%	1,400	1,531	+9%	0	0	—
case-05 We need to limit outgoing bandwidth on Linux interface eth0 to 1 Mbps, with a burst buffer size of 32 kbits and a maximum latency parameter of 400ms. An engineer suggested using netem rate limiting, but a Token Bucket Filter is required for strict rate enforcement. Provide the tc command to configure this rate limiting.	pass→pass	11,207	4,582	-59%	1	1	0%	1,202	1,513	+26%	0	0	—
case-06 We need to set up an isolated network testing environment on Linux using network namespaces. A colleague suggested assigning veth0 into the test namespace 'test_ns', but standard setup keeps veth0 in the root namespace and moves veth1 into 'test_ns'. Show the Linux commands to create the namespace 'test_ns', pair veth0 and veth1, move veth1 into test_ns, assign IP 10.0.0.2/24 to veth1 inside test_ns, and bring veth1 up in test_ns.	pass→pass	10,878	9,603	-12%	1	1	0%	1,371	1,665	+21%	0	0	—
case-07 We are simulating network chaos against a Redis server running on localhost:6380 by routing client traffic through a local proxy on localhost:6379. We need to create a Toxiproxy proxy named 'redis_proxy' and introduce a 1000ms latency toxic to it. Show the toxiproxy-cli commands to create the proxy and add this latency toxic.	pass→pass	5,003	8,944	+79%	1	1	0%	1,046	1,499	+43%	0	0	—
case-08 When setting up chaos engineering and network simulation scripts for a Node.js repository, we need to target specific entry points. A developer recommended targeting generic scripts like app-router.js. Which process targets are designated for network simulation and fuzzing testing in this stack?	fail→pass	20,935	8,105	-61%	1	1	0%	2,669	1,152	-57%	0	0	—
case-09 We need a specialized tool for performing container-level chaos engineering (such as dropping containers or injecting network delay into specific Docker containers). A team member recommended using standard iptables rules manually inside containers. What dedicated tool is specifically suited for container chaos testing?	pass→pass	16,843	14,059	-17%	1	1	0%	2,054	2,204	+7%	0	0	—
case-10 We need to dynamically inject network faults (latency, bandwidth throttling, connection drops) into TCP service connections during automated integration testing without requiring root privileges or tc commands. What dedicated tool provides a TCP proxy for chaos testing?	pass→pass	17,022	13,773	-19%	1	1	0%	1,984	2,169	+9%	0	0	—
case-11 When establishing automated quality gates for network simulation test suites, an engineer suggested that verifying application performance under degraded conditions is sufficient before declaring the test pass. What critical post-degradation verification phase must also be completed as a quality gate?	pass→pass	13,060	8,928	-32%	1	1	0%	1,314	1,972	+50%	0	0	—
case-12 Before running chaos scenarios and applying artificial latency or packet loss in network simulation tests, what benchmarking step must be completed to accurately evaluate the impact of degraded conditions?	pass→pass	11,734	11,487	-2%	1	1	0%	1,088	1,678	+54%	0	0	—
case-13 We are designing a WAN emulation test harness on Linux using netem. An engineer claims that netem can only simulate packet delay and packet loss. What additional WAN packet manipulation capabilities does netem support for corrupting or reordering traffic?	pass→pass	19,749	13,972	-29%	1	1	0%	2,733	3,097	+13%	0	0	—
case-14 We want to simulate a complex multi-switch, multi-host network topology locally for protocol testing without physical hardware or multiple virtual machines. What network emulation tool allows creating virtual network topologies with switches, hosts, and links?	pass→pass	9,632	11,009	+14%	1	1	0%	1,677	2,785	+66%	0	0	—
case-15 We need a simple command-line utility written in Go that wraps system tools like tc, iptables, and ipfw to easily simulate bad network connections (delay, loss, bandwidth limits) across Linux and macOS. What tool provides this cross-platform wrapper functionality?	pass→pass	14,835	6,717	-55%	1	1	0%	1,525	1,853	+22%	0	0	—
case-16 In our JavaScript-based test suite, we need to run heavy traffic simulation against our socket backend. Which target process file from our project targets is designated for load testing operations?	fail→pass	15,564	8,952	-42%	1	1	0%	1,898	944	-50%	0	0	—
case-17 We are setting up socket communication tests under network degradation. Which target process file in our test framework provides the TCP socket server endpoint for these tests?	fail→pass	4,329	6,833	+58%	1	1	0%	656	919	+40%	0	0	—
case-18 In our automated chaos pipeline, after issuing netem commands to simulate packet loss, what quality gate check must occur before running application assertions?	pass→pass	25,305	10,544	-58%	1	1	0%	1,636	1,531	-6%	0	0	—
case-19 When specifying latency with variance in a tc netem command (such as delay 100ms 20ms), what is the structural order between the baseline delay value and the jitter variance value?	pass→pass	5,488	6,318	+15%	1	1	0%	1,041	1,768	+70%	0	0	—
case-20 We are observing BGP route flapping on our edge routers (Cisco ASR 1000) running BGP AS 65001. How do we configure BGP route dampening parameters (half-life, reuse threshold, suppress threshold) on Cisco IOS-XE?	pass→pass	14,217	19,505	+37%	1	1	0%	2,741	3,386	+24%	0	0	—
case-21 We captured a packet trace in Wireshark (.pcap file) from a live production server. How do we construct a display filter expression in Wireshark to isolate HTTP GET requests containing a specific Authorization header token?	pass→pass	13,134	8,899	-32%	1	1	0%	1,559	2,284	+47%	0	0	—
case-22 We need to configure an Nginx reverse proxy block with round-robin load balancing across three upstream backend servers (10.0.0.10, 10.0.0.11, 10.0.0.12) with health check parameters. Provide the nginx.conf directives.	pass→pass	13,797	13,904	+1%	1	1	0%	1,834	2,043	+11%	0	0	—
case-23 Our microservices run on a Kubernetes cluster. How can container networking simulation tools be integrated into Docker or Kubernetes environments for testing microservice degradation?	pass→pass	25,567	23,255	-9%	1	1	0%	3,490	3,986	+14%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted, and 22 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +17 percentage points is the difference between those two pass rates over the 22 comparable cases.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 82.6%
Pass rate with the skill: 100%
Cases attempted: 23
Cases counted toward lift: 22
Cases that regressed: 0
Skill version measured: v1
Run completed: 8/30/2026, 08:58 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-08

Prompt

When setting up chaos engineering and network simulation scripts for a Node.js repository, we need to target specific entry points. A developer recommended targeting generic scripts like app-router.js. Which process targets are designated for network simulation and fuzzing testing in this stack?

Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.

Show both answers

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run100

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns+27% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 25 days ago

Other measured skills in the registry, with their headline benchmark lift.