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 automateyournetwork/netclaw --skill netbox-reconcilegit clone --depth 1 https://github.com/automateyournetwork/netclawWrote 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/automateyournetwork/netclaw/netbox-reconcile)<a href="https://agentmods.dev/skills/automateyournetwork/netclaw/netbox-reconcile"><img src="https://agentmods.dev/badge/skills/automateyournetwork/netclaw/netbox-reconcile/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/automateyournetwork/netclaw/netbox-reconcile"><img src="https://agentmods.dev/badge/skills/automateyournetwork/netclaw/netbox-reconcile.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00067 | $0.04759 |
| Opus 5 | $0.00034 | $0.02380 |
| Sonnet 5 | $0.00013 | $0.00952 |
| Haiku 4.5 | $0.00007 | $0.00476 |
Grade A, and why
netbox-reconcile 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 7d 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.
How it starts
The opening of the file, as written. The whole thing — 405 lines — stays where its author put it; the contents beside it link to each section on GitHub.
NetBox Reconciliation
Golden Rule
NetBox is READ-WRITE. The MCP has full API access to create and update devices, IPs, interfaces, VLANs, and cables. However, during reconciliation, discrepancies are reported and ticketed first — they are NEVER auto-corrected without explicit human approval. NetBox is the intended state. If reality differs from NetBox, either the network is wrong or NetBox needs updating. NetClaw can update NetBox when explicitly authorized by the operator.
How to Call the Tools
NetBox MCP Server
python3 $MCP_CALL "python3 -u $NETBOX_MCP_SCRIPT" TOOL_NAME '{"param":"value"}'
pyATS MCP Server (for live device state)
PYATS_TESTBED_PATH=$PYATS_TESTBED_PATH python3 $MCP_CALL "${PYATS_PYTHON:-python3} -u $PYATS_MCP_SCRIPT" TOOL_NAME '{"param":"value"}'
NetBox Tools Reference
| Tool | Purpose | Key Parameters |
|---|---|---|
netbox_get_objects |
Bulk query objects by type | object_type, filters (dict), limit, brief |
netbox_get_object_by_id |
Get single object by ID | object_type, object_id |
netbox_search_objects |
Global text search | query, object_types |
netbox_get_changelogs |
Audit trail of NetBox changes | filters |
Object Types
| Object Type | Contains |
|---|---|
dcim.devices |
Devices: name, role, platform, site, status |
dcim.interfaces |
Interfaces: name, type, enabled, MAC, MTU, mode (access/tagged), untagged_vlan, tagged_vlans |
ipam.ip-addresses |
IP addresses: address (CIDR), assigned_object, status, role |
dcim.cables |
Physical cables: A-side termination, B-side termination, type, length |
ipam.vlans |
VLANs: vid, name, site, tenant, status |
ipam.prefixes |
Prefixes: prefix (CIDR), VRF, site, status |
dcim.sites |
Sites: name, region, physical_address |
Discrepancy Types
| Code | Name | Severity | Description |
|---|---|---|---|
IP_DRIFT |
IP Address Drift | CRITICAL | Device IP differs from NetBox assignment |
MISSING_INTERFACE |
Missing Interface | HIGH | Interface exists in NetBox but not on device (or vice versa) |
UNDOCUMENTED_LINK |
Undocumented Link | HIGH | CDP/LLDP shows a neighbor connection not documented in NetBox cables |
CABLE_MISMATCH |
Cable Mismatch | HIGH | NetBox cable endpoints do not match CDP/LLDP neighbor data |
VLAN_MISMATCH |
VLAN Mismatch | MEDIUM | Device VLAN assignment differs from NetBox |
STATUS_MISMATCH |
Interface Status Mismatch | MEDIUM | NetBox shows enabled but device shows down (or vice versa) |
MTU_MISMATCH |
MTU Mismatch | LOW | Device MTU differs from NetBox interface MTU |
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.
- 7d ago First seen · 405 lines · 67 tokens per session scan A 9bb5d58f9733
netbox-reconcile is a skill published in the GitHub repository automateyournetwork/netclaw (655 stars, last pushed 4d ago), licensed Apache-2.0. It adds 67 tokens to every session and 4,759 once invoked, about $0.0003 per session on Opus 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 2026-09-03.
Other skills, from other repositories
equipment-fleet-manager
Manage construction equipment fleet. Track utilization, maintenance, and assignments.
tia-multiuser
TIA Portal Multiuser Engineering operations. Use for managing server projects, local sessions, and multiuser commissioning workflows.
bootstrap-asic-project
Bootstrap and coordinate a fresh open-source ASIC project when the PDK, EDA runtime, project configuration, or physical-design collateral is missing or ambiguous. Use for requests to choose an open PDK, provision compatible tools, create a reproducible ASIC workspace, turn RTL such as a CPU into a core or full-chip…
embedded-project-loop
Use first when Arduino or embedded hardware work spans sessions, parts delivery, wiring, soldering, flashing, measurements, field tests, recovery, or a durable next-todo is needed. Maintain explicit evidence and concrete physical-world user gates instead of assuming completion.
biomass-power-plant-manager
Use when a task needs the judgment of a Biomass Power Plant Manager — running a biomass-fired power generation facility, managing fuel supply logistics and quality variability, evaluating combustion/efficiency problems, or balancing generation output against fuel cost and emissions constraints.
ewm
Use when working with SAP Extended Warehouse Management — warehouse structure design, inbound/outbound processes, wave management, task and resource management, slotting, labor management, yard management, RF framework, VAS, or embedded EWM in S/4HANA.