> ## Documentation Index
> Fetch the complete documentation index at: https://docs.extractor.live/llms.txt
> Use this file to discover all available pages before exploring further.

# Transaction Monitor

> Match Solana transactions using ordered `conditions` records.

Match Solana transactions using ordered `conditions` records.

## Functionality

* Add one or more rows under **`conditions`**. Each row has three keys:

* **`field`**: which part of the transaction to test (`error`, `status`, `signer`, `accounts`, `fee`, `consumed`, `version`, `logs`).

* **`condition`**: the rule for that field (regex, comparison, exact value, or comma-separated accounts — see other FAQ entries).

* **`op`**: `OR` or `AND` — how this row combines with the running result from **previous** rows (left → right). The **first** row only seeds the accumulator; its `op` is ignored.

A row is skipped if `condition` is blank after trim. If `conditions` is empty, the detector does nothing.

* \| `field` | `condition` means |
  \| --- | --- |
  \| `error` | Must be a **valid** regex (optional leading `!` for negative). When `tx.err` is set, the pattern runs on it; when there is **no** error, only a **negative** regex matches (metadata value `null`). Invalid regex never matches. |
  \| `status` | If the string compiles as a regex (optional `!`), it is applied to `tx.sta`; if compilation **fails**, the whole string is compared to `tx.sta` with case-insensitive equality. |
  \| `signer` | Case-insensitive equality of `condition` to the **first** account key (fee payer). |
  \| `accounts` | Comma-separated addresses; match if **any** appears among message account keys (case-insensitive). |
  \| `fee` | Numeric comparison on `tx.fee` (e.g. `> 100000`, `== 5000`). |
  \| `consumed` | Numeric comparison on `tx.cost` (compute units consumed). |
  \| `version` | Exact match to `tx.ver`, or treat empty `ver` as `legacy` and match that. |
  \| `logs` | Valid regex on each log line; **any** line may satisfy. Invalid regex never matches. |
* **error** and **logs** use normal regex rules; prefix with **`!`** for a negative regex (match when the pattern does **not** match).

**status**: you can use a plain string (e.g. `Ok`) or a regex (e.g. `InstructionError`). If the string is not a valid regex, it is compared as exact text (case-insensitive).

Examples:

* `error` + `.*6001.*` — custom program error in `tx.err`.

* `error` + `!Success` — when you intend regex semantics on the error string (not the same as “successful tx” unless `tx.err` contains `Success`).

* `status` + `Ok` — successful transaction status.

* Use comparison operators on numeric fields:

* `>100000`, `<5000`, `>=10000`, `<=50000`

* `==5000` or `=5000`

* `!=0`

Whitespace around the operator and number is allowed.

* For **`logs`**, the regex is tested against **each** log line; one hit is enough. Examples: `Program.*failed`, `Error:.*`, `Program.*invoke`.

* When a transaction matches, metadata typically includes:

* **`conditions`**: comma-separated list of **field** names that matched.

* **`err`**, **`status`**, **`signer`**, **`accounts`**, **`fee`**, **`consumed`**, **`version`**, **`logs`**: values for rows that matched (each key appears only if that field matched).

* **`desc`**: from the config template.

The transaction signature is on the event’s transaction object (`tx.sig`), not necessarily in metadata.

## Configuration

* **Conditions** (`conditions`, default: `[{'op': 'AND', 'field': 'error', 'condition': 'InstructionError'}, {'op': 'AND', 'field': 'fee', 'condition': '>6500000'}]`): Ordered rules; each row: `field`, `condition`, `op` (see FAQ).
* **Description** (`desc`, default: `{conditions}: {error} {fee}`): Alert description template.
* **Severity** (`severity`, default: `-1`): Severity override (-1 = auto / default). See [common configuration](/common-configurations#severity).

<Note>
  Supported networks: solana.
</Note>