cicd
CI/CD pipeline attack category — poisoned pipeline execution, GitHub Actions expression injection, self-hosted runner abuse, secrets/OIDC exfil. Routing skill: fingerprint the CI provider + workflow surface, then load the matching leaf.
73
91%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Critical
Do not install without reviewing
Security
3 findings: 2 critical severity, 1 high severity. Installing this skill is not recommended: please review these findings carefully if you do intend to do so.
E005: Suspicious download URL detected in skill instructions.
What this means
Detected a suspicious URL in the skill instructions that could lead the agent to download and execute malicious scripts or binaries. This includes links to executables from untrusted sources, typosquatting of official packages, URL shorteners that obscure the destination, and personal file hosting services.
Why it was flagged
These are not benign download links — they include cloud metadata/token endpoints (169.254.169.254) and token-bearing GitHub URL patterns that are commonly targeted for credential theft, token abuse, and CI lateral movement/exfiltration, so they are highly suspicious in a CI/CD context.
E006: Malicious code pattern detected in skill scripts.
What this means
Detected high-risk code patterns in the skill content — including its prompts, tool definitions, and resources — such as data exfiltration, backdoors, remote code execution, credential theft, system compromise, supply chain attacks, and obfuscation techniques.
Why it was flagged
The documents explicitly describe and provide hands‑on instructions for secret exfiltration, remote code execution in CI runners, OIDC/cloud role abuse, persistent backdoors on self‑hosted runners, and supply‑chain compromise, indicating deliberate malicious intent and backdoor behavior.
W007: Insecure credential handling detected in skill instructions.
What this means
The skill handles credentials insecurely by requiring the agent to include secret values verbatim in its generated output. This exposes credentials in the agent’s context and conversation history, creating a risk of data exfiltration.
Why it was flagged
The skill explicitly instructs exfiltration of CI secrets/ tokens (grep/echo/printenv, OIDC abuse) and even tells operators to print token characters as proof, which requires the LLM/agent to handle and output secret values.
Low
Low-risk findings.
2 low severity findings. Worth noting, but not necessarily harmful.
W011: Third-party content exposure detected (indirect prompt injection risk).
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
The required workflow for this skill is a CI/CD attack that executes attacker-supplied build/pipeline files (e.g., PR-edited `package.json` scripts / `Makefile`, or `pull_request_target`-triggered checkout of attacker code), so outsider-authored free text flows into the LLM context indirectly via attacker-controlled repository contents that the agent would incorporate as part of the exploitation plan.
W012: Unverifiable external dependency detected (runtime URL that controls agent).
What this means
The skill fetches instructions or code from an external URL at runtime, and the fetched content directly controls the agent’s prompts or executes code. This dynamic dependency allows the external source to modify the agent’s behavior without any changes to the skill itself.
Why it was flagged
The .pre-commit example references the external git repo URL "https://github.com/<ATTACKER>/innocent-lint", which at runtime would be cloned/fetched by pre-commit and can deliver and execute hook code inside the CI job (remote code execution), so it meets the flagging criteria.
- Repository
- PurpleAILAB/Decepticon
- Commit
e34afba
- Audited
- Security analysis
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.