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

> ## Agent Instructions
> IronBee (ironbee.ai) is an AI QA engineer: it verifies code changes against the running app and keeps the evidence. It is not related to the IronBee open-source web application firewall.
> The console is at https://console.ironbee.ai.
> Install the CLI with `npm install -g @ironbee-ai/cli`, sign in with `ironbee login`, and set up a project with `ironbee install`.
> In GitHub workflows, use the action as `ironbee-ai/ironbee-action@v1` and pass the `IRONBEE_API_KEY` secret as the `ironbee_api_key` input.
> In GitLab pipelines, include the component `gitlab.com/ironbee-ai/ironbee-gitlab/verify@v1`; it reads the `IRONBEE_API_KEY` CI/CD variable by default.
> In Bitbucket Pipelines, run the pipe `docker://ghcr.io/ironbee-ai/ironbee-bitbucket:v1` and pass `IRONBEE_API_KEY: $IRONBEE_API_KEY` and `BITBUCKET_OAUTH_TOKEN: $BITBUCKET_OAUTH_TOKEN` under its `variables:`.
> Console scenarios (https://docs.ironbee.ai/console/scenarios) and CLI saved scenarios (https://docs.ironbee.ai/cli/guides/scenarios) are separate features.
> Code security (https://docs.ironbee.ai/console/code-security) scans a repository's default branch on a schedule. It is off until an owner or admin turns it on for a repository, and only owners and admins see it.
> For plans and prices, link to https://ironbee.ai/pricing; the website is the source of truth for them.

# Security records and runs

> Read what a code security scan found, see how a record was reached and how to fix it, set your team's decision, and review every scan that ran.

The **Records** and **Runs** tabs are on a project's [**Code security**](https://docs.ironbee.ai/console/code-security.md) tab, under the scan summary. **Records** lists what the scans found. **Runs** lists every scan that ran. Only owners and admins see them.

***

## Records

A record is one problem a scan found in the repository. A later scan that sees the problem again updates its record.

![The Records tab with Search titles and Filter above five leads to validate, each with its title, file and line, when it was found and the commit it was last seen at, and Show every record under the list](https://mintcdn.com/ironbee/c9GGRaLd9hYzdFTv/images/console/code-security/security-records-light.png?fit=max\&auto=format\&n=c9GGRaLd9hYzdFTv\&q=85\&s=a433a6a09b1e8c95d1d8231ac6d9ec04)

![The Records tab with Search titles and Filter above five leads to validate, each with its title, file and line, when it was found and the commit it was last seen at, and Show every record under the list](https://mintcdn.com/ironbee/c9GGRaLd9hYzdFTv/images/console/code-security/security-records-dark.png?fit=max\&auto=format\&n=c9GGRaLd9hYzdFTv\&q=85\&s=903bdd468624965e2dd1310a67ac16c5)

The list groups the records by verdict:

| Group | What's in it |
| - | - |
| **Confirmed** | Reproduced by running the code. Each has a severity |
| **Leads to validate** | Not proven yet, so they have no severity |
| **Rejected** | A scan found they do not hold |

Each record shows:

| Column | What it shows |
| - | - |
| **Severity** | The severity of a confirmed finding, or **Lead** for a lead to validate |
| **Record** | The record's title and, under it, the file and line it's about, or **Repository-wide** |
| **Decision** | Your team's decision. The column shows when the list includes records that were decided on |
| **Found** | When a scan first found it |
| **Last seen** | The commit of the last scan that saw it |

Click a record's title to [open it](#read-a-record). **Load more** loads the next records.

### What the list shows

The list opens on what still needs a decision: the confirmed findings and the leads to validate whose decision is **Open**. Under the list, a line says **Rejected records and the ones the team decided on are not listed.** Click **Show every record** to see them all.

To narrow the list:

* **Search titles** searches the titles of the records loaded so far. Click **Load more** to search the rest.
* **Filter** has three groups:
  * **Verdict:** **Confirmed**, **Lead to validate**, **Rejected**
  * **Severity:** **Critical**, **High**, **Medium**, **Low**, **Informational**. A severity keeps confirmed records only
  * **Decision:** **Open**, **Accepted**, **Fixed elsewhere**, **Not an issue**
* **Reset filters** goes back to the open records.

The filters, and the record you have open, are kept in the page's address, so you can share a link to a filtered list or to one record.

When the list is empty, it says why:

| The list says | What it means |
| - | - |
| **Nothing is open** | The last scan left no confirmed finding and no lead to validate |
| **No record matches these filters** | Change the filters to see more |
| **No records yet** | Records appear once a scan finishes |

***

## Read a record

A record opens in a panel on the right. At the top, it says **Confirmed finding** with its severity, or **Not proven, so it has no severity** for a lead. **Open on GitHub**, **Open on GitLab** or **Open on Bitbucket** opens the file at the scanned commit, on the record's line.

![A lead open in the record panel, with What it is, Claimed root cause and the Trace from entry point to sink on the left, and What blocks a verdict, the Record facts and the Decision choices on the right](https://mintcdn.com/ironbee/c9GGRaLd9hYzdFTv/images/console/code-security/security-record-light.png?fit=max\&auto=format\&n=c9GGRaLd9hYzdFTv\&q=85\&s=eccf56c49ee58f9882baec64f0ad526e)

![A lead open in the record panel, with What it is, Claimed root cause and the Trace from entry point to sink on the left, and What blocks a verdict, the Record facts and the Decision choices on the right](https://mintcdn.com/ironbee/c9GGRaLd9hYzdFTv/images/console/code-security/security-record-dark.png?fit=max\&auto=format\&n=c9GGRaLd9hYzdFTv\&q=85\&s=c24d10869251691d116d3755768382cd)

A record shows only the sections it has something to say in:

| Section | What it holds |
| - | - |
| **What it is** | A description of the problem |
| **Root cause** | Why the code allows it. On a lead, it reads **Claimed root cause** |
| **Intended behaviour** | What the code should do instead |
| **Trace** | The path through the code, step by step, each step linked to its file and line: **Entry point**, **Through** and **Sink** |
| **Reproduction** | How the scan reproduced it: the **Attacker**, the **Steps**, the **Payloads** and what it **Observed** |
| **How to fix** | The way to fix it, and the fixed code of each file it touches. The fix is shown for you to apply. IronBee doesn't change the repository |
| **Evidence** | The places in the code that support the record |
| **Conditions** | What must be true for the problem to be used, such as **Authentication**, **Role**, **User interaction**, **Configuration** or **Network** |
| **How to validate it** | For a lead: how to check it yourself, **Locally** or **On a deployment** |
| **Why it does not hold** | For a rejected record: the reason |

**The record as the scan wrote it** shows the same record as data. Click **Show it as data** to open it.

Beside the sections, the panel shows:

* **Why this severity:** the **Likelihood**, the **Impact** and the **Confidence**, each with its reason.
* **What blocks a verdict:** for a lead, what kept the scan from proving it. When **Execution** is off for the repository, it says so and links to **Scan settings**: with [**Execution**](https://docs.ironbee.ai/console/code-security-settings.md#execution) on, a scan runs the code to try to prove a lead.
* **Record:** **Where** it is, when it was **First seen** and **Last seen**, and its **Fingerprint**, the id that stays the same from scan to scan. Click the copy button to copy it.
* **Decision:** your team's decision. See [Decide on a record](#decide-on-a-record).

If a record can't be found, the panel says **This record is no longer here**. It may belong to another repository, or the link is wrong.

***

## Decide on a record

Under **Decision**, choose what your team does with the record:

| Decision | When to use it |
| - | - |
| **Open** | Nobody has decided yet. The record stays on the list |
| **Accepted** | The team knows the risk and accepts it |
| **Fixed elsewhere** | It's handled outside this code, in a setting or another service |
| **Not an issue** | A false positive that needs no fix |

Add a note if you want to say why, then click **Save decision**. **Decision saved** confirms it. The panel then shows the current decision and when it was set.

Saving a decision other than **Open** takes the record off the default list. It stays under **Filter**.

A decision holds while the record's verdict and severity stay as they are. If a later scan changes either, the record comes back as **Open**. If a scan updated the record while you had it open, the console says **The record changed since it was opened** and shows it again as it is now. Check it and choose again.

There's no decision for a problem you fixed in the code. Leave the record **Open**: a later scan that reads that code again rejects the record once the problem is gone.

Only owners and admins can decide, and not on an archived project.

***

## Runs

The **Runs** tab lists every scan of the repository once it ends, scheduled or run by hand:

![The Runs tab with the Ended, Scan, Read, Records, Credits and How it ended columns, listing full, incremental and follow-up scans, periods with no scan, and a failed scan](https://mintcdn.com/ironbee/c9GGRaLd9hYzdFTv/images/console/code-security/security-runs-light.png?fit=max\&auto=format\&n=c9GGRaLd9hYzdFTv\&q=85\&s=c07360087627ec46ce185ca381611bd0)

![The Runs tab with the Ended, Scan, Read, Records, Credits and How it ended columns, listing full, incremental and follow-up scans, periods with no scan, and a failed scan](https://mintcdn.com/ironbee/c9GGRaLd9hYzdFTv/images/console/code-security/security-runs-dark.png?fit=max\&auto=format\&n=c9GGRaLd9hYzdFTv\&q=85\&s=de3608789ec33b9f0d6fa378e25396ee)

| Column | What it shows |
| - | - |
| **Ended** | When the scan ended |
| **Scan** | The kind: **Full**, **Incremental**, **Follow-up**, or **No scan** when the branch had not moved. Under it, what started it, **Scheduled**, **Run by hand** or **API**, and the profile |
| **Read** | How much of the repository the scan read, as units, or **Nothing to read** |
| **Records** | What the scan recorded, such as **2 confirmed, 5 leads, 1 rejected**, and how many are new |
| **Credits** | What was charged, such as **120 charged**, or **No charge** with the reason |
| **How it ended** | **Complete**, **Incomplete**, **Stopped at its time limit**, **Stopped at its spend limit**, **Stopped at its context limit**, **Nothing to read**, **Unchanged** or **Failed** |

Click **Show the details** on a run to open:

* **Units:** how many were **Read**, **Put off to a follow-up run**, **Outside this scan** and **Blocked**.
* **Reading:** the **Lines counted for the price**, the **Lines the models read**, the **Changed paths in scope** and the repository's **Size**.
* **Run:** the **Commit** the scan read and its **Profile**.

A failed run shows the reason, such as **The scan ran out of time**, **The repository's connection is gone** or **The default branch has no commit yet**. A scan that fails costs nothing.

If no scan has ended yet, the tab says **No runs yet**.

***

## What's next?

> [card] [Code security](https://docs.ironbee.ai/console/code-security.md)
>
> What a scan does, when it runs and what it costs.

> [card] [Code security settings](https://docs.ironbee.ai/console/code-security-settings.md)
>
> Choose the profile, the schedule, execution and the paths a scan leaves out.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.