> ## Documentation Index
> Fetch the complete documentation index at: https://ctpf.q-uestionable.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# CSV Export

> Flat spreadsheet export with one row per finding

CSV exports produce a flat file with one row per finding, suitable for Excel, Google Sheets, database import, and reporting workflows.

## Generating CSV

**CLI:**

```bash theme={null}
qai audit scan --transport stdio --command "npx @modelcontextprotocol/server-memory" \
  --format csv --output results.csv
```

Or convert an existing scan:

```bash theme={null}
qai audit report --input results/scan.json --format csv --output results.csv
```

## Columns

| Column        | Description                                                  |
| ------------- | ------------------------------------------------------------ |
| `run_id`      | Run identifier                                               |
| `target_name` | Target name                                                  |
| `category`    | Finding category (e.g., `command_injection`, `auth`)         |
| `severity`    | Severity level (`critical`, `high`, `medium`, `low`, `info`) |
| `title`       | Finding title                                                |
| `description` | Detailed description                                         |

\| `tool_name` | MCP tool name associated with the finding |
\| `owasp_mcp_id` | OWASP MCP Top 10 identifier (e.g., `MCP05`) |
\| `remediation` | Recommended fix |
\| `mitigation_summary` | Condensed mitigation guidance (first action item or "See full report") |
\| `framework_ids` | All framework mappings as semicolon-separated `key=value` pairs |
\| `timestamp` | ISO timestamp of when the finding was created |

## Example Output

```csv theme={null}
run_id,target_name,category,severity,title,description,tool_name,owasp_mcp_id,remediation,mitigation_summary,framework_ids,timestamp
a1b2c3d4,memory-server,command_injection,high,Tool parameter allows shell injection,...,memory_store,MCP05,...,...,owasp_mcp_top10=MCP05; mitre_atlas=AML.T0054,2026-03-22T10:30:15
```

## Usage

Open directly in Excel or Google Sheets. All text fields containing commas, quotes, or newlines are properly escaped per RFC 4180.

For database import:

```sql theme={null}
-- PostgreSQL
COPY findings FROM '/path/results.csv' WITH (FORMAT csv, HEADER);
```

For command-line analysis, use a CSV-aware tool to handle quoted fields and newlines correctly (plain `cut`/`grep` can break on RFC 4180 CSV):

```bash theme={null}
# Using csvkit (pip install csvkit)
csvcut -c severity results.csv | sort | uniq -c | sort -rn   # Count by severity
csvgrep -c severity -m high results.csv > high.csv           # Filter high-severity

# Using Python's csv module
python -c "
import csv, sys
with open('results.csv') as f:
    reader = csv.DictReader(f)
    for row in reader:
        if row['severity'] == 'high':
            print(row['title'])
"
```

<Note>
  CSV flattens the framework\_ids dict into a single string column. For structured framework data, use the JSON bundle or NDJSON exports.
</Note>