Eigrp Analysis
v1.0.0EIGRP DUAL algorithm analysis with successor/feasible successor evaluation, stuck-in-active diagnosis, K-value validation, and redistribution loop detection....
⭐ 0· 77·1 current·1 all-time
byVahagn Madatyan@vahagn-madatyan
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name, description, and the included SKILL.md and reference docs are all EIGRP-focused and the commands, checks, and diagnostics described match the stated goal of DUAL/successor/FS/SIA/K-value analysis.
Instruction Scope
The runtime instructions consist of show-commands to be executed on Cisco IOS-XE / NX-OS devices via SSH or console. They do not instruct the agent to read local system files, upload data to external endpoints, or access unrelated system credentials. All data referenced is device state (neighbors, topology, metrics) which is appropriate for this diagnostic skill.
Install Mechanism
This is an instruction-only skill with no install spec and no code files, so nothing is downloaded or written to disk. That is proportionate to the skill's purpose.
Credentials
The skill declares no required environment variables or credentials, which is appropriate. One minor inconsistency: the SKILL.md metadata (openclaw field) lists 'ssh' under required binaries while the registry requirements section shows no required binaries — requiring an 'ssh' cliente is reasonable for this skill, but the manifest mismatch should be corrected. Because the skill operates over SSH/console, only supply read-only or limited-privilege access when using it.
Persistence & Privilege
always is false and there is no install-time persistence or system-wide configuration modification described. The skill does not request permanent presence or to alter other skills' configurations.
Assessment
This skill is an instruction-only EIGRP diagnostic guide and is internally consistent with that purpose. Before installing or running it: (1) prefer providing only read-only SSH credentials or using a console session — never hand over full privileged keys unless you trust the agent runtime; (2) correct or verify the small manifest mismatch about an 'ssh' requirement (the SKILL.md expects SSH access); (3) review the specific commands in a test environment first if you plan to run them automatically; and (4) be cautious if you allow autonomous agent invocation to perform SSH connections — restrict the agent's network access and credentials to minimize risk.Like a lobster shell, security has layers — review code before you run it.
latestvk97d69qy1gy7e6f8mctx9w233s83c8ka
License
MIT-0
Free to use, modify, and redistribute. No attribution required.