GitHub Actions Workflow Modification Blocked

ELASTIC detection rule by Elastic · Elastic License v2

Detects when a GitHub Actions workflow attempts to create or modify workflow files in a protected branch but is blocked due to insufficient permissions. This behavior is indicative of a supply chain attack where a malicious package or compromised CI/CD pipeline attempts to inject persistent backdoor workflows into a repository.

[metadata]
creation_date = "2025/12/05"
integration = ["github"]
maturity = "production"
updated_date = "2026/03/24"

[rule]
author = ["Elastic"]
description = """
Detects when a GitHub Actions workflow attempts to create or modify workflow files in a protected branch but is blocked
due to insufficient permissions. This behavior is indicative of a supply chain attack where a malicious package or
compromised CI/CD pipeline attempts to inject persistent backdoor workflows into a repository.
"""
false_positives = [
    """
    Legitimate CI/CD automation that requires workflow file modifications may trigger this alert if not properly
    configured with the necessary permissions. Review the workflow configuration and ensure the GITHUB_TOKEN or PAT has
    the required 'workflows' permission if the modification is intentional.
    """,
]
from = "now-9m"
interval = "8m"
language = "esql"
license = "Elastic License v2"
name = "GitHub Actions Workflow Modification Blocked"
note = """## Triage and analysis

### Investigating GitHub Actions Workflow Modification Blocked

This rule detects attempts to push workflow files to a GitHub repository from within a GitHub Actions workflow that are blocked by GitHub's security controls. This is a key indicator of supply chain attacks where malicious code attempts to establish persistence by injecting backdoor workflows.

### Possible investigation steps

- Review the `github.repo` field to identify which repository was targeted.
- Examine the `github.actor_id` to determine if the action was triggered by a bot (`github-actions[bot]`) or a user account (PAT-based).
- Check recent workflow runs in the repository for suspicious activity, especially in jobs that run `npm install` or other package manager commands.
- Review the repository's dependencies for recently added or updated packages that may contain malicious preinstall/postinstall hooks.
- Examine the `github.reasons.message` field for details on which workflow file was being created or modified.
- Search for other repositories in the organization that may have the same malicious dependency.
- Review GitHub audit logs for successful workflow file modifications that may have occurred before protections were enabled.

### False positive analysis

- Legitimate automation tools that manage workflow files may trigger this alert. Verify if the repository uses tools like Dependabot, Renovate, or custom automation that modifies workflows.
- CI/CD pipelines that intentionally update workflow files should use a PAT with the 'workflows' scope and be documented.

### Response and remediation

- If this is a confirmed attack attempt, immediately audit all dependencies in the affected repository.
- Remove any suspicious packages and regenerate lock files.
- Rotate any secrets that may have been exposed during the CI run.
- Review and revoke any PATs that may have been compromised.
- Enable branch protection rules requiring pull request reviews for workflow file changes.
- Consider implementing CODEOWNERS for `.github/workflows/` directory.
- Search for indicators of compromise such as unexpected workflow files (e.g., `discussion_*.yaml`, `formatter_*.yml`).
"""
references = ["https://www.wiz.io/blog/shai-hulud-2-0-ongoing-supply-chain-attack"]
risk_score = 47
rule_id = "e8b37f18-4804-4819-8602-4aba1169c9f4"
severity = "medium"
tags = [
    "Domain: Cloud",
    "Use Case: Threat Detection",
    "Tactic: Initial Access",
    "Tactic: Persistence",
    "Tactic: Execution",
    "Data Source: Github",
    "Resources: Investigation Guide",
]
timestamp_override = "event.ingested"
type = "esql"

query = '''
from logs-github.audit-* metadata _id, _index, _version
| where
    data_stream.dataset == "github.audit" and
    event.action == "protected_branch.rejected_ref_update" and
    github.category == "protected_branch" and
    github.reasons.code == "workflow_updates" and
    match(github.reasons.message::STRING, "refusing to allow a GitHub App to create or update workflow")
| keep *
'''


[[rule.threat]]
framework = "MITRE ATT&CK"

[[rule.threat.technique]]
id = "T1195"
name = "Supply Chain Compromise"
reference = "https://attack.mitre.org/techniques/T1195/"

[[rule.threat.technique.subtechnique]]
id = "T1195.001"
name = "Compromise Software Dependencies and Development Tools"
reference = "https://attack.mitre.org/techniques/T1195/001/"

[[rule.threat.technique.subtechnique]]
id = "T1195.002"
name = "Compromise Software Supply Chain"
reference = "https://attack.mitre.org/techniques/T1195/002/"

[rule.threat.tactic]
id = "TA0001"
name = "Initial Access"
reference = "https://attack.mitre.org/tactics/TA0001/"

[[rule.threat]]
framework = "MITRE ATT&CK"

[[rule.threat.technique]]
id = "T1059"
name = "Command and Scripting Interpreter"
reference = "https://attack.mitre.org/techniques/T1059/"

[rule.threat.tactic]
id = "TA0002"
name = "Execution"
reference = "https://attack.mitre.org/tactics/TA0002/"

[[rule.threat]]
framework = "MITRE ATT&CK"

[[rule.threat.technique]]
id = "T1546"
name = "Event Triggered Execution"
reference = "https://attack.mitre.org/techniques/T1546/"

[rule.threat.tactic]
id = "TA0003"
name = "Persistence"
reference = "https://attack.mitre.org/tactics/TA0003/"

[[ currentRule.title && currentRule.title.length > 60 ? currentRule.title.slice(0, 60) + '…' : currentRule.title ]]

76bad0e9-96fc-41fc-8c28-a768b381b1a9

Propose edits, view related rules, and engage with the community through comments.

Executable code, not a declarative rule

This submission is arbitrary [[ currentRule.format ? currentRule.format.toUpperCase() : '' ]] code, not a declarative detection rule — review it as you would any script before running it, and note that it cannot be safely aggregated or auto-executed alongside YARA/Sigma/Suricata/CRS rules.

Rejected pattern present

[[ reason ]]

What does this mean?

Format & documentation suggestions

Not a security issue — this rule is valid and was accepted as-is. These are optional fields/conventions its own format recommends, mainly affecting its quality score and how well it documents itself for other reviewers.

[[ reason ]]

What do these messages mean?

[[ currentRule.title && currentRule.title.length > 60 ? currentRule.title.slice(0, 60) + '…' : currentRule.title ]]

[[ currentRule.title ]]

Community Discussion