Skip to content
Skills Library
Data & analytics · Analysis

network-data-analysis

/network-data-analysis

Ad-hoc read-only SQL analysis over exported network data (Zeek logs, Suricata eve.json, generated reports) using DuckDB. Use when aggregating across a packet capture's sessions, correlating IDS alerts with connection metadata, or answering counting and grouping questions that a per-log view cannot

Repository
netclaw ↗
Licence
Apache-2.0
Last repository change
25.10.2026

Assessment and security

100/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content35 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
1
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 25.10.2026, without references to retired models or APIs.

Licence

Apache-2.0, otwarta: można używać i zmieniać, z zachowaniem licencji i informacji o zmianach.

What it does and when to use it

Ad-hoc read-only SQL analysis over exported network data (Zeek logs, Suricata eve.json, generated reports) using DuckDB. Use when aggregating across a packet capture's sessions, correlating IDS alerts with connection metadata, or answering counting and grouping questions that a per-log view cannot

Original author description

Ad-hoc read-only SQL analysis over exported network data (Zeek logs, Suricata eve.json, generated reports) using DuckDB. Use when aggregating across a packet capture's sessions, correlating IDS alerts with connection metadata, or answering counting and grouping questions that a per-log view cannot

How it works

  1. MCP Server
  2. What this reads, and what it can never read
  3. Workflow: analyse a capture
  4. The pivot that makes this worth using
  5. Reading results honestly
  6. Important Rules
  7. Integration with Other Skills
  8. Environment Variables
Original SKILL.md headings · 106 lines, 701 words
SKILL.md preview
---
name: network-data-analysis
description: "Ad-hoc read-only SQL analysis over exported network data (Zeek logs, Suricata eve.json, generated reports) using DuckDB. Use when aggregating across a packet capture's sessions, correlating IDS alerts with connection metadata, or answering counting and grouping questions that a per-log view cannot"
version: 1.0.0
license: Apache-2.0
tags: [analysis, sql, duckdb, zeek, suricata, forensics, read-only]
---

# Network Data Analysis (read-only SQL)

## MCP Server

- **Server**: `analysis-mcp` (NetClaw-authored, spec 092)
- **Tools**: `analysis_status`, `analysis_datasets`, `analysis_query`
- **Engine**: DuckDB, in-memory, **sandboxed and locked**

## What this reads, and what it can never read

Loads files from an **allowlist of roots** — `~/.openclaw/nsm/runs` (Zeek/Suricata output
from `nsm-mcp`), the workspace output directory, and `~/.openclaw/analysis` for your own
exports.

**NetClaw's own stores are permanently unreachable**: `~/.openclaw/memory/`,
`~/.openclaw/rag/`, `~/.openclaw/n2n/` and `~/.openclaw/gait/`, plus `.ssh`, `.aws`, `.kube`
and `.env`. A generic SQL surface over those would be a backdoor, not an analysis tool.

That is not enforced by pattern matching. Datasets are materialised, then DuckDB's own
`enable_external_access=false` and `lock_configuration=true` close every filesystem and
network path **irreversibly** for the life of the process. Verified: `read_csv('/etc/passwd')`,
`glob('/home/**')`, `ATTACH` of the memory or RAG stores, `COPY … TO`, `INSTALL`/`LOAD`, and
re-enabling access all raise.

If you need a file analysed, put it under `~/.openclaw/analysis` — do not try to reach it in
SQL, because you cannot.

## Workflow: analyse a capture

1. **Produce the data first.** `nsm_analyze` (skill `nsm-session-pivot`) writes Zeek logs and
   Suricata `eve.json`. Without a run, this surface has nothing to read and says so.
2. `analysis_status` — is the sandbox locked, how many datasets loaded, what are the caps?
3. `analysis_datasets` — table names, row counts, columns. **Read the columns before writing
   SQL**; guessing column names wastes a round trip.
4. `analysis_query` — one read statement per call.

Zeek tables carry their **real column names** (`id_orig_h`, `id_resp_p`, `uid`), lifted from
the log's `#fields` header. Without that they would be `column0…columnN` and unusable.

## The pivot that makes this worth using

Every Zeek log shares `uid` with `conn.log`, so a join *is* the session pivot — and unlike
walking logs one at a time, it aggregates:

```sql
SELECT c.id_orig_h, c.id_resp_h, c.service, h.method, h.host, h.uri
FROM zeek_<run>_conn c
LEFT JOIN zeek_<run>_http h USING (uid)
ORDER BY c.ts
```

Counting questions a per-log view cannot answer:

```sql
-- top talkers by connection count
SELECT id_orig_h, count(*) AS conns FROM zeek_<run>_conn
GROUP BY 1 ORDER BY conns DESC LIMIT 20

-- which services appeared at all
SELECT service, count(*) FROM zeek_<run>_conn GROUP BY 1 ORDER BY 2 DESC

-- Suricata alert signatures by frequency
SELECT json_extract_string(alert, '$.signature') AS sig, count(*)
FROM suricata_<run>_eve WHERE event_type = 'alert' GROUP BY 1 ORDER BY 2 DESC
```

## Reading results honestly

- **`truncated: true` means you are looking at a page.** Never present a capped result as a
  total — run `COUNT(*)` for the real number. The tool reports this in `gaps`.
- **Zeek columns are all text.** Datasets load as varchar so a malformed field cannot abort
  the load; cast explicitly (`CAST(duration AS DOUBLE)`) and say that you did.
- **A row count reflects what was LOADED, not what existed.** Files above the size cap are
  skipped and listed in `notes`; a per-table row cap applies. Check `analysis_status` before
  claiming completeness.
- **`0 datasets` means no exports exist**, never "the network was quiet."
- **This inherits every caveat of its source.** A Zeek run made with checksum validation on
  may be missing whole protocol logs (see `nsm-session-pivot`) — and SQL over an incomplete
  log is confidently wrong. Check the posture of the run you are querying.

## Important Rules

- **Read-only.** `INSERT`/`UPDATE`/`DELETE`/`DROP`/`CREATE`/`ATTACH`/`COPY`/`INSTALL`/`SET`
  are refused, and independently impossible after lockdown.
- **One statement per call.** A stacked second statement is refused.
- **Queries time out** (default 30s) and are interrupted, not left running.
- **Record in GAIT** — log the query and the dataset it ran against, not just the answer.

## Integration with Other Skills

| Skill | How They Work Together |
|-------|----------------------|
| `nsm-session-pivot` | Produces the Zeek logs this queries; use it for single-session detail |
| `nsm-ids-triage` | Produces `eve.json`; use SQL here to aggregate alerts across a capture |
| `packet-analysis` | Drop to individual packet decode once SQL narrows the field |
| `document-generation` | Turn a query result into a report table |
| `gait-session-tracking` | Record all analysis runs |

## Environment Variables

- `ANALYSIS_QUERY_TIMEOUT` — per-query seconds (default 30)
- `ANALYSIS_MAX_RESULT_ROWS` — result cap (default 500)
- `ANALYSIS_EXTRA_ROOTS` — extra allowlisted roots, `os.pathsep`-separated
- `ANALYSIS_MAX_FILE_BYTES` / `ANALYSIS_MAX_ROWS` — load caps

Versions

The database has no other skills named network-data-analysis.

Identical content appears in 1 other GitHub folders. We show one copy with the highest score.

Database snapshot: 13.11.2026 · package version 58204738ed52 (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 100ids-deployment-build
    Deploy and operate a network IDS sensor with Suricata (optionally alongside Zeek) — capture design (SPAN/TAP, af-packet), HOME_NET and variables, rule sources with suricata-update, custom local rules, EVE JSON output to the SIEM, performance tuning, health monitoring — verified…
  • 100suricata-alert-triage
    Triage Suricata/Snort/Zeek-notice IDS alerts using eve.json fields, the signature's logic, flow and app-layer metadata, payload/PCAP evidence and host corroboration to decide true/false positive and whether an attack succeeded. Use when working network IDS alerts in a SOC or…
  • 100elasticsearch-logs
    Search and analyse logs in an existing Elasticsearch cluster (8.x/9.x) — syslog, application logs, Zeek/Suricata exports, or any indexed event data. Use for "what errors did we see", "how many times did X happen", "show me logs from device Y", "which host logged the most".…
  • 95duckdb-en
    DuckDB CLI specialist for SQL analysis, data processing and file conversion. Use for SQL queries, CSV/Parquet/JSON analysis, database queries, or data conversion. Triggers on "duckdb", "sql", "query", "data analysis", "parquet", "convert data".
  • 90nsm-ids-triage
    Triage Suricata IDS alerts from a packet capture (read-only) — signature alerts with detection posture, severity filtering, and corroboration against Zeek session metadata. Use when triaging IDS alerts, checking whether a capture contains known-bad traffic, or validating that a…
  • 100configuring-suricata-for-network-monitoring
    Deploys and configures Suricata IDS/IPS with Emerging Threats rulesets, EVE JSON logging, and custom rules for high-throughput, protocol-aware traffic inspection (HTTP, TLS, DNS, SMB) and SIEM integration. Use when running Suricata in IDS or inline IPS mode to detect or block…

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.