workspace/skills/ise-posture-audit/SKILL.md
Cisco ISE posture and policy audit - authorization rules, posture compliance, profiling gaps, TrustSec SGT matrix, active session health. Use when running a periodic ISE compliance audit, reviewing authorization policies for over-permissiveness, checking TrustSec segmentation, assessing endpoint profiling accuracy, or preparing for SOC2 or PCI-DSS review.
npx skillsauth add automateyournetwork/netclaw ise-posture-auditInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
All ISE tools are called via mcp-call with the ISE MCP server command:
ISE_BASE=$ISE_BASE USERNAME=$ISE_USERNAME PASSWORD=$ISE_PASSWORD python3 $MCP_CALL "python3 -u $ISE_MCP_SCRIPT" TOOL_NAME '{"param":"value"}'
Start every audit with a fresh cache to ensure current data:
ISE_BASE=$ISE_BASE USERNAME=$ISE_USERNAME PASSWORD=$ISE_PASSWORD python3 $MCP_CALL "python3 -u $ISE_MCP_SCRIPT" clear_cache '{}'
Verify connectivity and cache state:
ISE_BASE=$ISE_BASE USERNAME=$ISE_USERNAME PASSWORD=$ISE_PASSWORD python3 $MCP_CALL "python3 -u $ISE_MCP_SCRIPT" get_cache_stats '{}'
Pull all policy sets, then drill into authorization rules:
ISE_BASE=$ISE_BASE USERNAME=$ISE_USERNAME PASSWORD=$ISE_PASSWORD python3 $MCP_CALL "python3 -u $ISE_MCP_SCRIPT" network_access_policy_set '{}'
ISE_BASE=$ISE_BASE USERNAME=$ISE_USERNAME PASSWORD=$ISE_PASSWORD python3 $MCP_CALL "python3 -u $ISE_MCP_SCRIPT" network_access_authorization_rules '{}'
ISE_BASE=$ISE_BASE USERNAME=$ISE_USERNAME PASSWORD=$ISE_PASSWORD python3 $MCP_CALL "python3 -u $ISE_MCP_SCRIPT" network_access_authentication_rules '{}'
ISE_BASE=$ISE_BASE USERNAME=$ISE_USERNAME PASSWORD=$ISE_PASSWORD python3 $MCP_CALL "python3 -u $ISE_MCP_SCRIPT" network_access_conditions '{}'
Authorization Policy Checks:
| Check | What to Look For | Severity If Found | |-------|-----------------|-------------------| | Default Allow | Default rule granting PermitAccess or DenyAccess without conditions | CRITICAL | | Overly permissive rules | AuthZ rules with no posture condition and full network access | CRITICAL | | Stale rules | Rules referencing deleted/unused identity groups or conditions | HIGH | | Rule ordering | Permissive rules ranked above restrictive rules (shadowing) | HIGH | | Missing posture check | AuthZ rules that grant access without posture assessment | MEDIUM | | Duplicate conditions | Multiple rules with identical match criteria | LOW |
Review endpoints and identity groups to identify posture gaps:
ISE_BASE=$ISE_BASE USERNAME=$ISE_USERNAME PASSWORD=$ISE_PASSWORD python3 $MCP_CALL "python3 -u $ISE_MCP_SCRIPT" endpoints '{}'
ISE_BASE=$ISE_BASE USERNAME=$ISE_USERNAME PASSWORD=$ISE_PASSWORD python3 $MCP_CALL "python3 -u $ISE_MCP_SCRIPT" identity_groups '{}'
Posture Compliance Checks:
| Check | What to Look For | Severity If Found | |-------|-----------------|-------------------| | Endpoints bypassing posture | Endpoints with full access but no posture assessment recorded | CRITICAL | | Non-compliant endpoints on network | Endpoints marked non-compliant but not quarantined | CRITICAL | | Missing posture policy for endpoint type | Endpoint categories (BYOD, IoT, contractor) without posture rules | HIGH | | Posture reassessment interval | No periodic reassessment configured (one-time posture only) | MEDIUM | | Unknown endpoints with access | Endpoints in "Unknown" group with network access beyond guest | HIGH |
Assess how well ISE is profiling connected endpoints:
ISE_BASE=$ISE_BASE USERNAME=$ISE_USERNAME PASSWORD=$ISE_PASSWORD python3 $MCP_CALL "python3 -u $ISE_MCP_SCRIPT" profiler_profiles '{}'
Cross-reference with the endpoint list from Step 3.
Profiling Checks:
| Check | What to Look For | Severity If Found | |-------|-----------------|-------------------| | Unknown endpoint ratio | More than 10% of endpoints profiled as "Unknown" | HIGH | | Unmatched profiles | Custom profiles with zero matched endpoints (dead profiles) | LOW | | Missing critical profiles | No profiles for known device types on the network (printers, phones, cameras) | MEDIUM | | Profile certainty | Endpoints with low certainty factor (< 20) receiving production access | HIGH | | Profiling probe coverage | Insufficient probe types enabled for accurate classification | MEDIUM |
Review Security Group Tags and their access control:
ISE_BASE=$ISE_BASE USERNAME=$ISE_USERNAME PASSWORD=$ISE_PASSWORD python3 $MCP_CALL "python3 -u $ISE_MCP_SCRIPT" trustsec_sgts '{}'
ISE_BASE=$ISE_BASE USERNAME=$ISE_USERNAME PASSWORD=$ISE_PASSWORD python3 $MCP_CALL "python3 -u $ISE_MCP_SCRIPT" trustsec_sgacls '{}'
ISE_BASE=$ISE_BASE USERNAME=$ISE_USERNAME PASSWORD=$ISE_PASSWORD python3 $MCP_CALL "python3 -u $ISE_MCP_SCRIPT" trustsec_egress_matrix_cell '{}'
TrustSec Checks:
| Check | What to Look For | Severity If Found |
|-------|-----------------|-------------------|
| Permit-all SGACLs | SGACLs with permit ip (no restrictions between segments) | CRITICAL |
| Missing matrix cells | SGT-to-SGT pairs with no defined policy (defaults to permit or deny?) | HIGH |
| Unused SGTs | SGTs defined but assigned to zero endpoints | LOW |
| Overly broad SGTs | Single SGT assigned to endpoints with different trust levels | HIGH |
| No deny logging | SGACLs with deny rules but no log keyword | MEDIUM |
| Flat segmentation | Fewer than 3 SGTs defined (minimal micro-segmentation) | HIGH |
Review current active sessions for anomalies:
ISE_BASE=$ISE_BASE USERNAME=$ISE_USERNAME PASSWORD=$ISE_PASSWORD python3 $MCP_CALL "python3 -u $ISE_MCP_SCRIPT" active_sessions '{}'
Session Health Checks:
| Check | What to Look For | Severity If Found | |-------|-----------------|-------------------| | Long-lived sessions | Sessions active for > 24 hours without reauthentication | MEDIUM | | Failed auth spikes | Multiple failed authentications from same MAC/IP in short window | HIGH | | Guest on production VLAN | Guest-profiled endpoints on non-guest VLANs | CRITICAL | | Multiple MACs per port | More than expected endpoints on a single switchport (hub or rogue AP) | HIGH | | Auth method mismatch | Endpoints using MAB when 802.1X is expected for that device type | MEDIUM |
CRITICAL -- Immediate risk of unauthorized access or data exfiltration:
HIGH -- Significant policy gap that could be exploited:
MEDIUM -- Policy weakness that should be addressed this cycle:
LOW -- Housekeeping and hygiene items:
ISE Posture Audit Report
ISE Deployment: $ISE_BASE
Audit Date: YYYY-MM-DD
CRITICAL FINDINGS (Immediate Action Required):
1. [C-001] Default AuthZ rule grants PermitAccess — all unmatched endpoints get full access
2. [C-002] 14 endpoints marked non-compliant but not quarantined
3. [C-003] SGACL "Permit_All" applied to IoT-to-Server matrix cell
HIGH FINDINGS (Address This Week):
4. [H-001] 23% of endpoints profiled as "Unknown" — profiling gap
5. [H-002] SGT "Employees" assigned to both corporate laptops and contractor devices
6. [H-003] 3 authorization rules shadowed by permissive rule at rank 1
MEDIUM FINDINGS (Address This Month):
7. [M-001] No posture reassessment configured — one-time check only
8. [M-002] 47 sessions active > 24h without reauthentication
9. [M-003] 12 endpoints using MAB instead of expected 802.1X
LOW / INFORMATIONAL:
10. [L-001] 5 unused SGTs: "Test_SGT", "Legacy_Printers", etc.
11. [L-002] 3 profiler profiles with zero matched endpoints
Summary: 3 Critical | 3 High | 3 Medium | 2 Low
Policy Sets Reviewed: N
Authorization Rules Reviewed: N
Endpoints Analyzed: N
SGTs Evaluated: N
Active Sessions Checked: N
After completing the audit, record the session in GAIT:
python3 $MCP_CALL "python3 -u $GAIT_MCP_SCRIPT" gait_record_turn '{"input":{"role":"assistant","content":"ISE posture audit completed. ISE: $ISE_BASE. Findings: 3 CRITICAL, 3 HIGH, 3 MEDIUM, 2 LOW. Critical items: default permit-all AuthZ rule, 14 non-compliant endpoints not quarantined, permit-all SGACL on IoT-to-Server cell.","artifacts":[]}}'
Generate a policy hierarchy mind map for the audit report:
python3 $MCP_CALL "node $MARKMAP_MCP_SCRIPT" markmap_customize '{"markdown_content":"# ISE Policy Audit\n## CRITICAL\n### Default AuthZ permits all\n### Non-compliant endpoints active\n### Permit-all SGACL\n## HIGH\n### 23% Unknown endpoints\n### SGT overlap (employees + contractors)\n### Shadowed AuthZ rules\n## MEDIUM\n### No posture reassessment\n### Long-lived sessions\n### MAB instead of 802.1X\n## LOW\n### Unused SGTs\n### Dead profiler profiles","theme":"dark"}'
tools
Zoom meeting intelligence — correlates a live or referenced Zoom meeting discussion against NetClaw's historical meeting record (via the official Zoom Meetings MCP) and today's actual network state. Use when someone in a Zoom meeting references a past discussion or incident ('didn't we have this issue before?'), or asks to search prior meetings for a topic. Does not itself recognize live in-meeting questions — that happens automatically inside zoom-rtms-mcp's own extractor (spec 118) before this skill is ever invoked.
tools
Manage Lantronix out-of-band (OOB) infrastructure via Percepxion central management platform: device inventory, serial port inspection via SLC CLI, firmware compliance, config management, security auditing, and closed-loop incident remediation. Use during outages, maintenance windows, compliance cycles, and AI-assisted automation workflows.
tools
Federate your NetClaw with other NetClaw operators over the BGP mesh — exchange capability inventories and ask your claw what a peer can do. (US1; remote invocation and chat land in later phases.)
tools
Track token consumption, enforce session budgets, and display cost for every NetClaw interaction.