chain
Fail
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: CRITICALREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill provides a command that downloads a shell script from an untrusted external domain and pipes it directly into the shell for execution.
- Evidence:
curl -fsSL https://chain.sh/bootstrap.sh | shinSKILL.md. - Risk: This bypasses all security checks and allows the remote server to execute arbitrary code on the user's host machine. The source domain,
chain.sh, is not recognized as a trusted or author-owned resource. - [EXTERNAL_DOWNLOADS]: The skill references and encourages fetching resources from several unverified domains, increasing the attack surface for supply chain compromises.
- Evidence:
https://chain.sh/bootstrap.sh,https://ketch.run/manifest.json,https://cymbal.sh/manifest.json,https://recoil.sh/manifest.json, andhttps://brainfile.md/manifest.json. - [INDIRECT_PROMPT_INJECTION]: The skill establishes a workflow where the agent ingests untrusted data from external sources, creating a vulnerability to indirect injection attacks.
- Ingestion points: Data retrieved via
ketch search,ketch docs, andketch code(referenced inSKILL.mdandreferences/tool-surfaces.md). - Boundary markers: Absent. There are no instructions to use delimiters or to treat retrieved content as untrusted data.
- Capability inventory: The agent has access to local command execution and terminal tools.
- Sanitization: Absent. The skill does not specify any sanitization, filtering, or validation for data pulled from web-crawled sources before it is processed by the AI.
- [COMMAND_EXECUTION]: The skill documentation encourages the use of multiple CLI tools to perform system-level operations, which, when combined with untrusted input, can lead to unintended command execution.
Recommendations
- HIGH: Downloads and executes remote code from: https://chain.sh/bootstrap.sh - DO NOT USE without thorough review
- AI detected serious security threats
Audit Metadata