Teams should first preserve and isolate the CI system, then collect immutable evidence from the source repository, runner host, orchestration layer and central log platform. Missing run numbers usually mean the attacker altered or deleted job metadata, so compare commit history, pipeline definitions, artefact timestamps and authentication events to rebuild the execution chain before credentials are rotated or runners are rebuilt.
Why log disappearance after a CI compromise is an evidence-preservation problem first
When workflow logs vanish, the immediate problem is not just missing visibility, it is that the attacker may have altered the execution trail itself. The first job is to stop further change, preserve the CI environment, and capture evidence from every place the pipeline left traces, including the source repository, runner host, orchestration layer, and central log platform.
That order matters because logs are only one view of the run. Commit history, pipeline definitions, artefact timestamps, queue records, and authentication events can still show what executed even when the visible run record is incomplete or deleted.
Teams should treat the missing run number as a clue, not a nuisance, because it often points to tampering with job metadata, not a harmless logging failure. The goal is to reconstruct the execution chain before the environment is rebuilt or secrets are changed, so the investigation is based on evidence rather than on an attacker-controlled console.
What evidence rebuilds the execution chain when job metadata is missing
Start with the repository and pipeline definition that existed at the time of compromise. Compare commit history, branch changes, workflow file edits, and pull request activity to the pipeline configuration that was actually deployed, then line that up with artefact creation times and any runner-side execution records.
Next, correlate those records with authentication and control-plane events. CI compromise often leaves traces in audit logs, token use, runner registration, build orchestration, and log forwarding, so you want to confirm who or what submitted the job, which runner executed it, and whether any management action deleted or altered the trace.
If the central log platform is also suspect, rely on preserved copies and immutable storage rather than the live dashboard alone. A wiped or delayed log stream can hide the sequence of events, while side-channel evidence such as artefact timestamps and repository metadata can still reveal the sequence of the build.
Why credential rotation and rebuilds come after triage, not before it
Once the evidence picture is stable, teams can decide what to rotate, revoke, or rebuild. That sequence is important because immediate rotation may destroy the very artefacts needed to determine whether the attacker only touched logging or also accessed secrets, runners, or deployment paths.
After the initial preservation step, the practical question becomes blast radius: did the compromise stay inside the pipeline control plane, or did it reach credentials, signing material, deployment tokens, or downstream systems? The answer determines whether you can cleanly reissue access or whether you need a wider incident reset.
For teams that need a deeper response pattern for compromised pipelines, the relevant playbook is CI/CD Pipeline Compromise Playbook, which focuses on triage, evidence, disabling workflows and runners, and clean recovery. If the compromise includes missing or altered telemetry, the same response logic overlaps with If your log pipeline (collectors, forwarders, data pipeline) is compromised, because both problems require backfill, integrity checks, and careful containment.
Risk and Threat Considerations
Log disappearance after a CI compromise is dangerous because it can conceal the attacker’s movement, delay detection, and leave teams rebuilding the wrong incident timeline. The practical risk is not just loss of observability, but loss of confidence in what actually ran, what data moved, and which credentials or deployment paths were exposed.
Failure mechanism: An attacker with pipeline control can delete or alter job metadata, suppress log forwarding, or tamper with the systems that collect and retain audit evidence, which breaks the normal execution trail and makes later reconstruction harder.
Impact: Teams may rotate the wrong credentials first, miss early compromise indicators, or rebuild the environment before preserving proof of abuse, which increases the chance of repeated access, incomplete containment, and disputed incident scope.
Standards & Framework Alignment
This section maps relevant standards and security frameworks to the operational risks and controls described in this guidance.
MITRE ATT&CK addresses the attack and risk surface, while NIST SP 800-53 Rev 5 and CIS Controls v8 set the governance and control requirements practitioners need to meet.
| Framework | Control / Reference | Relevance |
|---|---|---|
| NIST SP 800-53 Rev 5 | AU-9 — Protection of Audit Information | CI log loss directly implicates audit record protection and immutability. |
| AU-6 — Audit Review, Analysis, and Reporting | The answer centers on correlating audit evidence across repo, runner, and platform logs. | |
| IR-4 — Incident Handling | The question asks what to do first during an active CI compromise response. | |
| Recommendation — Protect audit logs from deletion or alteration and preserve independent copies for incident reconstruction. Correlate audit events across CI, repository, runner, and log platforms to reconstruct the execution chain. Preserve evidence and contain the affected CI environment before remediation or rebuild. | ||
| CIS Controls v8 | 8 — Audit Log Management | Missing workflow logs are an audit-log integrity and retention problem. |
| Recommendation — Centralize, protect, and retain CI audit logs so tampering and deletion are detectable. | ||
| MITRE ATT&CK | T1562 — Impair Defenses | Deleting or suppressing workflow logs is a defense-impairment technique used to hide activity. |
| Recommendation — Hunt for log suppression and other defense-impairment activity around the CI compromise. | ||
Practitioner Guidance
What to prioritise: Preserve the CI system and its evidence sources before you remediate the environment. That means freezing changes, capturing repository state, runner state, orchestration data, and central log exports before any rebuild or secret reset changes the record.
What to verify: Confirm whether the missing logs are isolated to the visible UI or whether the underlying artefacts, audit events, and authentication records are also missing or altered. If the execution chain can be rebuilt from independent sources, treat the log loss as a tampering signal rather than a simple retention issue.
Decision rule: If you still cannot answer which code ran, on which runner, and under which identity, do not move straight to reconstruction. Keep collecting and correlating evidence until that chain is established, then decide whether rotation, runner rebuild, or broader containment is warranted.
Practitioner takeaway: In a CI compromise, the first objective is evidential integrity, not cleanup, because every response choice after that depends on whether the attacker merely obscured the trail or also used it to move deeper.
Related resources from NHI Mgmt Group
- What should security teams review first after an agentic workflow compromise?
- Who is accountable when exposed secrets are found in public workflow logs after a supply chain compromise?
- What should teams do in the first 24 to 72 hours after suspected package compromise?
- What should teams do in the first 24 to 72 hours after a connected app compromise?
Deepen Your Knowledge
Free weekly newsletter
Subscribe to the NHI & AI Identity Journal
The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.
Bonus 33% off our NHI Course when you subscribe.
Reviewed and updated by the NHIMG editorial team on November 16, 2026.
NHI Mgmt Group — the #1 independent authority on Non-Human Identity, IAM, and Agentic AI security. nhimg.org