The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
feat: salvage network diagnostics skills (#1731)
Affaan Mustafad2ade249f666
diff --git a/.claude-plugin/marketplace.json b/.claude-plugin/marketplace.jsonindex 6e9b770fe..b213aeb5c 100644--- a/.claude-plugin/marketplace.json+++ b/.claude-plugin/marketplace.json@@ -11,7 +11,7 @@{"name": "ecc","source": "./",- "description": "The most comprehensive Claude Code plugin — 53 agents, 195 skills, 69 legacy command shims, selective install profiles, and production-ready hooks for TDD, security scanning, code review, and continuous learning",+ "description": "The most comprehensive Claude Code plugin — 53 agents, 198 skills, 69 legacy command shims, selective install profiles, and production-ready hooks for TDD, security scanning, code review, and continuous learning","version": "2.0.0-rc.1","author": {"name": "Affaan Mustafa",diff --git a/.claude-plugin/plugin.json b/.claude-plugin/plugin.jsonindex 0673fbaf8..1b9b6f327 100644--- a/.claude-plugin/plugin.json+++ b/.claude-plugin/plugin.json@@ -1,7 +1,7 @@{"name": "ecc","version": "2.0.0-rc.1",- "description": "Battle-tested Claude Code plugin for engineering teams — 53 agents, 195 skills, 69 legacy command shims, production-ready hooks, and selective install workflows evolved through continuous real-world use",+ "description": "Battle-tested Claude Code plugin for engineering teams — 53 agents, 198 skills, 69 legacy command shims, production-ready hooks, and selective install workflows evolved through continuous real-world use","author": {"name": "Affaan Mustafa","url": "https://x.com/affaanmustafa"diff --git a/AGENTS.md b/AGENTS.mdindex f3a9852cd..9ac4e1492 100644--- a/AGENTS.md+++ b/AGENTS.md@@ -1,6 +1,6 @@# Everything Claude Code (ECC) — Agent Instructions-This is a **production-ready AI coding plugin** providing 53 specialized agents, 195 skills, 69 commands, and automated hook workflows for software development.+This is a **production-ready AI coding plugin** providing 53 specialized agents, 198 skills, 69 commands, and automated hook workflows for software development.**Version:** 2.0.0-rc.1@@ -146,7 +146,7 @@ Troubleshoot failures: check test isolation → verify mocks → fix implementat```agents/ — 53 specialized subagents-skills/ — 195 workflow skills and domain knowledge+skills/ — 198 workflow skills and domain knowledgecommands/ — 69 slash commandshooks/ — Trigger-based automationsrules/ — Always-follow guidelines (common + per-language)diff --git a/README.md b/README.mdindex 99b0cfce6..4d5037cb9 100644--- a/README.md+++ b/README.md@@ -350,7 +350,7 @@ If you stacked methods, clean up in this order:/plugin list ecc@ecc```-**That's it!** You now have access to 53 agents, 195 skills, and 69 legacy command shims.+**That's it!** You now have access to 53 agents, 198 skills, and 69 legacy command shims.### Dashboard GUI@@ -1338,7 +1338,7 @@ The configuration is automatically detected from `.opencode/opencode.json`.|---------|-------------|----------|--------|| Agents | PASS: 53 agents | PASS: 12 agents | **Claude Code leads** || Commands | PASS: 69 commands | PASS: 31 commands | **Claude Code leads** |-| Skills | PASS: 195 skills | PASS: 37 skills | **Claude Code leads** |+| Skills | PASS: 198 skills | PASS: 37 skills | **Claude Code leads** || Hooks | PASS: 8 event types | PASS: 11 events | **OpenCode has more!** || Rules | PASS: 29 rules | PASS: 13 instructions | **Claude Code leads** || MCP Servers | PASS: 14 servers | PASS: Full | **Full parity** |@@ -1443,7 +1443,7 @@ ECC is the **first plugin to maximize every major AI coding tool**. Here's how e|---------|------------|------------|-----------|----------|| **Agents** | 53 | Shared (AGENTS.md) | Shared (AGENTS.md) | 12 || **Commands** | 69 | Shared | Instruction-based | 31 |-| **Skills** | 195 | Shared | 10 (native format) | 37 |+| **Skills** | 198 | Shared | 10 (native format) | 37 || **Hook Events** | 8 types | 15 types | None yet | 11 types || **Hook Scripts** | 20+ scripts | 16 scripts (DRY adapter) | N/A | Plugin hooks || **Rules** | 34 (common + lang) | 34 (YAML frontmatter) | Instruction-based | 13 instructions |diff --git a/README.zh-CN.md b/README.zh-CN.mdindex 0373aee54..1702fa540 100644--- a/README.zh-CN.md+++ b/README.zh-CN.md@@ -160,7 +160,7 @@ Copy-Item -Recurse rules/typescript "$HOME/.claude/rules/"/plugin list ecc@ecc```-**完成!** 你现在可以使用 53 个代理、195 个技能和 69 个命令。+**完成!** 你现在可以使用 53 个代理、198 个技能和 69 个命令。### multi-* 命令需要额外配置diff --git a/docs/zh-CN/AGENTS.md b/docs/zh-CN/AGENTS.mdindex 434bea5bc..d52859fe0 100644--- a/docs/zh-CN/AGENTS.md+++ b/docs/zh-CN/AGENTS.md@@ -1,6 +1,6 @@# Everything Claude Code (ECC) — 智能体指令-这是一个**生产就绪的 AI 编码插件**,提供 53 个专业代理、195 项技能、69 条命令以及自动化钩子工作流,用于软件开发。+这是一个**生产就绪的 AI 编码插件**,提供 53 个专业代理、198 项技能、69 条命令以及自动化钩子工作流,用于软件开发。**版本:** 2.0.0-rc.1@@ -147,7 +147,7 @@```agents/ — 53 个专业子代理-skills/ — 195 个工作流技能和领域知识+skills/ — 198 个工作流技能和领域知识commands/ — 69 个斜杠命令hooks/ — 基于触发的自动化rules/ — 始终遵循的指导方针(通用 + 每种语言)diff --git a/docs/zh-CN/README.md b/docs/zh-CN/README.mdindex 0d8938bc3..84413ffa0 100644--- a/docs/zh-CN/README.md+++ b/docs/zh-CN/README.md@@ -224,7 +224,7 @@ Copy-Item -Recurse rules/typescript "$HOME/.claude/rules/"/plugin list ecc@ecc```-**搞定!** 你现在可以使用 53 个智能体、195 项技能和 69 个命令了。+**搞定!** 你现在可以使用 53 个智能体、198 项技能和 69 个命令了。***@@ -1134,7 +1134,7 @@ opencode|---------|-------------|----------|--------|| 智能体 | PASS: 53 个 | PASS: 12 个 | **Claude Code 领先** || 命令 | PASS: 69 个 | PASS: 31 个 | **Claude Code 领先** |-| 技能 | PASS: 195 项 | PASS: 37 项 | **Claude Code 领先** |+| 技能 | PASS: 198 项 | PASS: 37 项 | **Claude Code 领先** || 钩子 | PASS: 8 种事件类型 | PASS: 11 种事件 | **OpenCode 更多!** || 规则 | PASS: 29 条 | PASS: 13 条指令 | **Claude Code 领先** || MCP 服务器 | PASS: 14 个 | PASS: 完整 | **完全对等** |@@ -1242,7 +1242,7 @@ ECC 是**第一个最大化利用每个主要 AI 编码工具的插件**。以|---------|------------|------------|-----------|----------|| **智能体** | 53 | 共享 (AGENTS.md) | 共享 (AGENTS.md) | 12 || **命令** | 69 | 共享 | 基于指令 | 31 |-| **技能** | 195 | 共享 | 10 (原生格式) | 37 |+| **技能** | 198 | 共享 | 10 (原生格式) | 37 || **钩子事件** | 8 种类型 | 15 种类型 | 暂无 | 11 种类型 || **钩子脚本** | 20+ 个脚本 | 16 个脚本 (DRY 适配器) | N/A | 插件钩子 || **规则** | 34 (通用 + 语言) | 34 (YAML 前页) | 基于指令 | 13 条指令 |diff --git a/manifests/install-modules.json b/manifests/install-modules.jsonindex e226cc0f8..054312b95 100644--- a/manifests/install-modules.json+++ b/manifests/install-modules.json@@ -512,9 +512,12 @@"kind": "skills","description": "Deployment workflows, Docker patterns, and infrastructure skills.","paths": [+ "skills/cisco-ios-patterns","skills/deployment-patterns","skills/docker-patterns","skills/homelab-network-setup",+ "skills/netmiko-ssh-automation",+ "skills/network-bgp-diagnostics","skills/network-config-validation","skills/network-interface-health"],diff --git a/package.json b/package.jsonindex df790d8cd..088e63b17 100644--- a/package.json+++ b/package.json@@ -104,6 +104,7 @@"skills/brand-voice/","skills/carrier-relationship-management/","skills/claude-devfleet/",+ "skills/cisco-ios-patterns/","skills/clickhouse-io/","skills/code-tour/","skills/coding-standards/",@@ -187,6 +188,8 @@"skills/messages-ops/","skills/nanoclaw-repl/","skills/nestjs-patterns/",+ "skills/netmiko-ssh-automation/",+ "skills/network-bgp-diagnostics/","skills/network-config-validation/","skills/network-interface-health/","skills/nodejs-keccak256/",diff --git a/skills/cisco-ios-patterns/SKILL.md b/skills/cisco-ios-patterns/SKILL.mdnew file mode 100644index 000000000..22f9840a2--- /dev/null+++ b/skills/cisco-ios-patterns/SKILL.md@@ -0,0 +1,163 @@+---+name: cisco-ios-patterns+description: Cisco IOS and IOS-XE review patterns for show commands, config hierarchy, wildcard masks, ACL placement, interface hygiene, and safe change-window verification.+origin: community+---++# Cisco IOS Patterns++Use this skill when reviewing Cisco IOS or IOS-XE snippets, building a+change-window checklist, or explaining how to collect evidence from a router or+switch without making the incident worse.++## When to Use++- Reviewing IOS or IOS-XE configuration before a planned change.+- Choosing read-only `show` commands for troubleshooting.+- Checking ACL wildcard masks and interface direction.+- Explaining global, interface, routing process, and line configuration modes.+- Verifying that a change landed in running config and was saved intentionally.++## Operating Rules++Treat IOS examples as patterns, not paste-ready production changes. Confirm the+platform, interface names, current config, rollback path, and out-of-band access+before making changes on a real device.++Prefer this workflow:++1. Capture current state with read-only commands.+2. Review the exact candidate config.+3. Confirm management access cannot be locked out.+4. Apply the smallest change in a maintenance window.+5. Re-read state, compare to the baseline, then save only after validation.++## Mode Reference++```text+Router> enable+Router# show running-config+Router# configure terminal+Router(config)# interface GigabitEthernet0/1+Router(config-if)# description UPLINK-TO-CORE+Router(config-if)# no shutdown+Router(config-if)# exit+Router(config)# end+Router# show running-config interface GigabitEthernet0/1+```++`running-config` is active memory. `startup-config` is what survives reload.+Do not save a change just because a command was accepted; validate behavior+first, then use `copy running-config startup-config` if the change is approved.++## Read-Only Collection++```text+show version+show inventory+show processes cpu sorted+show memory statistics+show logging+show running-config | section line vty+show running-config | section interface+show running-config | section router bgp+show ip interface brief+show interfaces+show interfaces status+show vlan brief+show mac address-table+show spanning-tree+show ip route+show ip protocols+show ip access-lists+show route-map+show ip prefix-list+```++Collect the specific section you need instead of dumping full config into a+ticket when the config may contain secrets, customer names, or private topology.++## Wildcard Masks++IOS ACL and many routing statements use wildcard masks, not subnet masks.++```text+Subnet mask Wildcard mask+255.255.255.255 0.0.0.0+255.255.255.252 0.0.0.3+255.255.255.0 0.0.0.255+255.255.0.0 0.0.255.255+```++Review wildcard masks before deployment. A subnet mask accidentally used as a+wildcard can match far more traffic than intended.++```text+ip access-list extended WEB-IN+ 10 permit tcp 192.0.2.0 0.0.0.255 any eq 443+ 999 deny ip any any log+```++Every ACL has an implicit deny at the end. Add an explicit logged deny when the+operational goal includes observing misses, and confirm logging volume is safe.++## ACL Placement Review++Before applying an ACL to an interface, answer these questions:++- Which traffic direction is being filtered, `in` or `out`?+- Is management traffic sourced from a known jump host or management subnet?+- Is there an explicit permit for required routing, DNS, NTP, monitoring, or+ application traffic?+- Are hit counters available from a safe test source?+- Is there a rollback command and an active console or out-of-band path?++Do not test reachability by removing firewall or ACL protections. Read counters,+logs, and route state first.++## Interface Hygiene++```text+interface GigabitEthernet0/1+ description UPLINK-TO-CORE+ switchport mode trunk+ switchport trunk allowed vlan 10,20,30+ switchport trunk native vlan 999+ no shutdown+```++Use clear descriptions, explicit switchport mode, and documented native VLANs.+On routed interfaces, confirm the mask, peer addressing, and routing process+before assuming link state means forwarding is correct.++## Change-Window Verification++Use before/after checks that match the actual change.++```text+show running-config | section interface GigabitEthernet0/1+show interfaces GigabitEthernet0/1+show logging | include GigabitEthernet0/1|changed state|line protocol+show ip route <prefix>+show ip access-lists <name>+```++For routing changes, also capture neighbor state and route tables before and+after the change. For ACL changes, compare hit counters from a planned test+source rather than relying on a generic ping.++## Anti-Patterns++- Applying a generated config without a device-specific diff.+- Saving configuration before post-change checks pass.+- Using a subnet mask where IOS expects a wildcard mask.+- Applying an ACL to the wrong interface direction.+- Troubleshooting by disabling ACLs, route policies, or authentication.+- Pasting full configs into public tools without sanitizing secrets and topology.++## See Also++- Agent: `network-config-reviewer`+- Agent: `network-troubleshooter`+- Skill: `network-config-validation`+- Skill: `network-interface-health`diff --git a/skills/netmiko-ssh-automation/SKILL.md b/skills/netmiko-ssh-automation/SKILL.mdnew file mode 100644index 000000000..1374ff53a--- /dev/null+++ b/skills/netmiko-ssh-automation/SKILL.md@@ -0,0 +1,173 @@+---+name: netmiko-ssh-automation+description: Safe Python Netmiko patterns for read-only collection, bounded batch SSH, TextFSM parsing, guarded config changes, timeouts, and network automation error handling.+origin: community+---++# Netmiko SSH Automation++Use this skill when writing or reviewing Python automation that connects to+network devices with Netmiko. Keep the default path read-only; config changes+need a separate change window, peer review, and rollback plan.++## When to Use++- Collecting `show` command output across routers, switches, or firewalls.+- Building a small audit script for interface, routing, or config evidence.+- Adding timeouts and exception handling to network SSH scripts.+- Parsing command output with TextFSM when a template exists.+- Reviewing automation before it touches production devices.++## Safety Defaults++- Start with read-only `send_command()` collection.+- Keep inventory small and explicit; do not sweep whole address ranges.+- Use environment variables, a vault, or `getpass`; never hardcode credentials.+- Set connection and read timeouts.+- Limit concurrency so older devices are not overloaded.+- Require an explicit operator flag before `send_config_set()`.+- Do not call `save_config()` until the change has been verified and approved.++## Read-Only Connection Pattern++```python+import os+from getpass import getpass+from netmiko import ConnectHandler+from netmiko.exceptions import (+ NetmikoAuthenticationException,+ NetmikoTimeoutException,+ ReadTimeout,+)++device = {+ "device_type": "cisco_ios",+ "host": "192.0.2.10",+ "username": os.environ.get("NETMIKO_USERNAME") or input("Username: "),+ "password": os.environ.get("NETMIKO_PASSWORD") or getpass("Password: "),+ "secret": os.environ.get("NETMIKO_ENABLE_SECRET"),+ "conn_timeout": 10,+ "auth_timeout": 20,+ "banner_timeout": 15,+ "read_timeout_override": 30,+}++try:+ with ConnectHandler(**device) as conn:+ if device.get("secret") and not conn.check_enable_mode():+ conn.enable()+ output = conn.send_command("show ip interface brief", read_timeout=30)+ print(output)+except NetmikoAuthenticationException:+ print("Authentication failed")+except NetmikoTimeoutException:+ print("SSH connection timed out")+except ReadTimeout:+ print("Command read timed out")+```++Use placeholder addresses from documentation ranges in examples. Keep real+inventory in an ignored local file or a secrets-managed system.++## Batch Collection++```python+from concurrent.futures import ThreadPoolExecutor, as_completed+from typing import Any++def collect_show(device: dict[str, Any], command: str) -> dict[str, Any]:+ host = device["host"]+ try:+ with ConnectHandler(**device) as conn:+ output = conn.send_command(command, read_timeout=45)+ return {"host": host, "ok": True, "output": output}+ except (NetmikoAuthenticationException, NetmikoTimeoutException, ReadTimeout) as exc:+ return {"host": host, "ok": False, "error": type(exc).__name__}++results = []+with ThreadPoolExecutor(max_workers=8) as pool:+ futures = [pool.submit(collect_show, device, "show version") for device in devices]+ for future in as_completed(futures):+ results.append(future.result())+```++Keep `max_workers` low unless the device estate and AAA systems are known to+handle higher connection volume.++## Structured Parsing++Netmiko can ask TextFSM, TTP, or Genie to parse supported command output. Treat+parser output as an optimization, not the only evidence path.++```python+with ConnectHandler(**device) as conn:+ parsed = conn.send_command(+ "show ip interface brief",+ use_textfsm=True,+ raise_parsing_error=False,+ read_timeout=30,+ )++if isinstance(parsed, str):+ print("No parser template matched; store raw output for review")+else:+ for row in parsed:+ print(row)+```++If parsing drives a blocking decision, keep the raw command output alongside+the parsed result so an operator can inspect mismatches.++## Guarded Config Pattern++```python+import os++commands = [+ "interface GigabitEthernet0/1",+ "description CHANGE-1234 UPLINK-TO-CORE",+]++apply_changes = os.environ.get("APPLY_NETWORK_CHANGES") == "1"++if not apply_changes:+ print("Dry run only. Candidate commands:")+ print("\n".join(commands))+else:+ with ConnectHandler(**device) as conn:+ conn.enable()+ before = conn.send_command("show running-config interface GigabitEthernet0/1")+ output = conn.send_config_set(commands)+ after = conn.send_command("show running-config interface GigabitEthernet0/1")+ print(before)+ print(output)+ print(after)+ print("Verify behavior before saving startup config.")+```++Saving the config is a separate approval step. In production, include a rollback+snippet and capture before/after evidence in the change record.++## Review Checklist++- Does the script identify an explicit inventory source?+- Are credentials absent from source, logs, and exception messages?+- Are `conn_timeout`, `auth_timeout`, and command `read_timeout` set?+- Are failures reported per device without stopping the whole batch?+- Does the script avoid broad scans and unbounded concurrency?+- Are config changes behind a dry-run or explicit operator flag?+- Is `save_config()` separate from the initial push and tied to verification?++## Anti-Patterns++- Hardcoding passwords, enable secrets, or private keys in source.+- Sending config commands as the default code path.+- Running automation against a CIDR range instead of a reviewed inventory.+- Logging full running configs to shared systems without sanitization.+- Treating parser success as proof that the device state is correct.++## See Also++- Skill: `cisco-ios-patterns`+- Skill: `network-config-validation`+- Skill: `network-interface-health`diff --git a/skills/network-bgp-diagnostics/SKILL.md b/skills/network-bgp-diagnostics/SKILL.mdnew file mode 100644index 000000000..d5c7eecfd--- /dev/null+++ b/skills/network-bgp-diagnostics/SKILL.md@@ -0,0 +1,167 @@+---+name: network-bgp-diagnostics+description: Diagnostics-only BGP troubleshooting patterns for neighbor state, route exchange, prefix policy, AS path inspection, and safe evidence collection.+origin: community+---++# Network BGP Diagnostics++Use this skill when a BGP session is down, flapping, established with missing+routes, or advertising unexpected prefixes. The default workflow is read-only+evidence collection; policy and reset actions belong in a reviewed change+window.++## When to Use++- BGP neighbors are stuck in Idle, Connect, Active, OpenSent, or OpenConfirm.+- A session is Established but expected prefixes are missing.+- A route-map, prefix-list, max-prefix limit, or AS path policy may be filtering+ routes.+- You need before/after evidence for a BGP change.+- You are reviewing automation that parses BGP summary output.++## Read-Only Triage Flow++1. Identify the exact neighbor, address family, VRF, and local/remote ASNs.+2. Capture summary state and last reset reason.+3. Prove reachability to the peer source address.+4. Check route policy references before assuming transport failure.+5. Compare advertised, received, and installed routes where the platform+ supports those commands.++```text+show bgp summary+show bgp neighbors <peer>+show ip route <peer>+show tcp brief | include <peer>|:179+show logging | include BGP|<peer>+show running-config | section router bgp+show ip prefix-list+show route-map+```++Use platform-specific address-family commands when the device uses VRFs, IPv6,+VPNv4, or EVPN. Do not assume global IPv4 unicast.++## State Interpretation++| State | First checks |+| --- | --- |+| Established with prefix count | Route exchange is up; inspect policy and table selection |+| Established with zero prefixes | Check inbound policy, max-prefix, advertised routes, and AFI/SAFI |+| Active | TCP session is not completing; check routing, source, ACLs, and peer reachability |+| Connect | TCP connection is in progress; check path and remote listener |+| OpenSent/OpenConfirm | TCP works; check ASN, authentication, timers, capabilities, and logs |+| Idle | Neighbor may be disabled, missing config, blocked by policy, or backoff timer |++## Transport Checks++```text+ping <peer> source <local-source>+traceroute <peer> source <local-source>+show ip route <peer>+show bgp neighbors <peer> | include BGP state|Last reset|Local host|Foreign host+```++If the peer is sourced from a loopback, confirm both directions route to the+loopback addresses and that the neighbor config uses the expected update source.++Avoid disabling ACLs or firewall policy as a diagnostic shortcut. Read hit+counters, logs, and path state first.++## Route Policy Checks++```text+show bgp neighbors <peer> advertised-routes+show bgp neighbors <peer> routes+show ip prefix-list <name>+show route-map <name>+show bgp <prefix>+```++Some platforms require additional configuration before `received-routes` is+available. Do not add that configuration during incident triage unless the+operator approves the change.++## AS Path And Prefix Review++```text+show bgp regexp _65001_+show bgp regexp ^65001$+show bgp <prefix>+show bgp neighbors <peer> advertised-routes | include Network|Path|<prefix>+```++Use AS-path regex carefully. `_65001_` matches AS 65001 as a token. Plain+`65001` can match longer ASNs or unrelated text.++## Parser Pattern++```python+import re+from typing import Any++BGP_SUMMARY_RE = re.compile(+ r"^(?P<neighbor>\d{1,3}(?:\.\d{1,3}){3})\s+"+ r"(?P<version>\d+)\s+"+ r"(?P<remote_as>\d+)\s+"+ r"(?P<msg_rcvd>\d+)\s+"+ r"(?P<msg_sent>\d+)\s+"+ r"(?P<table_version>\d+)\s+"+ r"(?P<input_queue>\d+)\s+"+ r"(?P<output_queue>\d+)\s+"+ r"(?P<uptime>\S+)\s+"+ r"(?P<state_or_prefixes>\S+)$",+ re.M,+)++def parse_bgp_summary(raw: str) -> list[dict[str, Any]]:+ rows = []+ for match in BGP_SUMMARY_RE.finditer(raw):+ state_or_prefixes = match.group("state_or_prefixes")+ if state_or_prefixes.isdigit():+ state = "Established"+ prefixes_received = int(state_or_prefixes)+ else:+ state = state_or_prefixes+ prefixes_received = None+ rows.append({+ "neighbor": match.group("neighbor"),+ "remote_as": int(match.group("remote_as")),+ "state": state,+ "prefixes_received": prefixes_received,+ "uptime": match.group("uptime"),+ })+ return rows+```++Prefer structured parser output when available, but store raw output with the+incident record because BGP summary formats vary by platform and address family.++## Change-Window Only++These actions can affect routing and should not be suggested as automatic+diagnostics:++- Clearing a BGP session.+- Changing neighbor authentication, timers, update source, route-maps, or+ prefix-lists.+- Enabling additional received-route storage.+- Relaxing firewall, ACL, or control-plane policy.++If a reset is approved, prefer the least disruptive soft or route-refresh option+supported by the platform and document exactly why it is safe.++## Anti-Patterns++- Assuming `Active` always means the remote side is down.+- Ignoring VRF, address family, or update-source differences.+- Using broad AS-path regex without token boundaries.+- Hard-resetting a peer before reading last reset reason and logs.+- Treating missing `received-routes` output as proof that no routes arrived.++## See Also++- Skill: `cisco-ios-patterns`+- Skill: `network-config-validation`+- Skill: `network-interface-health`