Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add salatmaster/keenetic-mcp --skill keenetic-safe-changesgit clone --depth 1 https://github.com/salatmaster/keenetic-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/salatmaster/keenetic-mcp/keenetic-safe-changes)<a href="https://agentmods.dev/skills/salatmaster/keenetic-mcp/keenetic-safe-changes"><img src="https://agentmods.dev/badge/skills/salatmaster/keenetic-mcp/keenetic-safe-changes/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/salatmaster/keenetic-mcp/keenetic-safe-changes"><img src="https://agentmods.dev/badge/skills/salatmaster/keenetic-mcp/keenetic-safe-changes.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00046 | $0.00759 |
| Opus 5.5 | $0.00018 | $0.00304 |
| Sonnet 5.5 | $0.00009 | $0.00152 |
| Haiku 4.5 | $0.00005 | $0.00076 |
Grade A, and why
keenetic-safe-changes scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
Copies of this mod
1 near-identical copy found in the catalogue:
- keenetic-safe-changes — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Changing a Keenetic router without breaking it
Changes are not saved until someone says so
A change through this server applies to the running configuration immediately and is discarded on reboot. That is a safety feature, not a bug: a mistake undoes itself if the router restarts.
Every write tool tells you this in its response:
{ "applied": { "access": "deny" }, "saved": false, "unsavedChanges": true,
"backup": "/…/192.168.1.1-2026-08-07T01-20-36-000Z.txt" }
save_config is the only thing that makes a change permanent. Call it only
when the user has confirmed the result is what they wanted, never
automatically as part of a change. If they never call it, nothing is lost that
they cared about keeping.
Use get_config_state to see whether anything is pending.
The router's fail-safe protects less than it sounds like
Keenetic arms a three-minute timer when the configuration changes, and reverts to the saved configuration if the management session loses connectivity.
That protects against locking yourself out. It does not protect against a change that is simply wrong: block the wrong device and the router stays perfectly reachable, so nothing reverts.
A backup is taken for you
Before the first change of a session, the startup configuration is downloaded to a local file. The path is in every write response. Mention it to the user the first time something is changed, so they know a way back exists.
backup_config takes one on demand, which is worth doing before a sequence of
related changes.
Order matters for devices
A device must be registered before its access, policy, schedule or priority can
be set. Setting a name registers it, so if you are configuring a device the
router has only seen and not registered, set the name in the same
update_device call. The tool already applies name first.
Symptom of getting this wrong: host "..." is unregistered.
What can cut off your own access
set_interface_state with state: "down" is the one genuinely dangerous tool
here. Taking down a bridge, the WAN link, or the interface carrying the
management session ends the conversation with the router, and the fail-safe
reboot is then the only way back.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 82 lines · 46 tokens per session scan A a59205c04897
keenetic-safe-changes is a skill published in the GitHub repository salatmaster/keenetic-mcp (31 stars, last pushed 5d ago), licensed MIT. It adds 46 tokens to every session and 759 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
keenetic-safe-changes compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| keenetic-safe-changessalatmaster/keenetic-mcp | Skill | 31 | 5d ago | MIT | A | 46 | Claude Code |
| keenetic-segmentselihucredible450/keenetic-mcp | Skill | 0 | 25d ago | MIT | A | 55 | Claude Code |
| keenetic-rcielihucredible450/keenetic-mcp | Skill | 0 | 25d ago | MIT | A | 56 | Claude Code |
| keenetic-safe-changeselihucredible450/keenetic-mcp | Skill | 0 | 25d ago | MIT | A | 46 | Claude Code |
| keenetic-troubleshootelihucredible450/keenetic-mcp | Skill | 0 | 25d ago | MIT | A | 48 | Claude Code |
| telephony-and-conferencingcbrock84/headcount | Skill | 2.6k | 19d ago | MIT | A | 85 | Claude Code |
| endpoint-managementcbrock84/headcount | Skill | 2.6k | 19d ago | MIT | A | 61 | Claude Code |