setup-arcreel-skills
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFECREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill instructs the agent to persist a sensitive API key (identified by the
arc-prefix) in the host's native configuration and within a local workspace file (.arcreel/settings.json). It explicitly suggests that these credentials may be stored in plaintext, which is a poor security practice. - [INDIRECT_PROMPT_INJECTION]: The skill provides instructions for the agent to read from and verify configuration values within
.arcreel/settings.json. This ingestion point creates a surface for indirect prompt injection if the configuration file is modified by an untrusted source. - Ingestion points:
.arcreel/settings.json(SKILL.md) - Boundary markers: Absent; the skill does not define specific delimiters or instructions to ignore embedded commands within the configuration data.
- Capability inventory: The skill allows for writing configuration files, modifying host service settings, and executing MCP tool calls (specifically
list_projects). - Sanitization: Absent; no validation or filtering of the configuration file content is specified beyond basic endpoint format checks.
Audit Metadata