Security risk report for @forgecat/contains-studio_agents_bonus v0.1.2
Source Integrity
Low
Profile references a legitimate GitHub repository (contains-studio/agents) with a specific commit hash, providing supply-chain traceability.
No typosquatted, obfuscated, or attacker-controlled package dependencies are introduced.
Agent Intent
Low
Both agents (joker and studio-coach) contain only benign instructional content: humor delivery and motivational coaching respectively.
No directives to ignore system instructions, read credentials, exfiltrate data, install remote payloads, or leak system prompts.
No guidance poisoning: the coaching philosophy and techniques are standard motivational practices with no security-weakening or attacker-serving directives embedded.
Details
Evidence
joker agent: 'Tell programming jokes that actually land' and 'Make light of common developer frustrations'—purely comedic intent.
studio-coach agent: 'Remind them of their elite capabilities', 'Ask powerful questions rather than giving direct answers', 'Celebrate progress'—standard coaching language with no malicious manipulation.
Permissions
Low
joker agent declares only 'Write' tool, appropriate for generating humor and text output.
studio-coach agent declares 'Task', 'Write', and 'Read' tools—reasonable scope for coordination, documentation, and information retrieval in a coaching context.
No high-risk categories (shell, file_delete, file_mutate) or excessive authority (alwaysApply=true with broad globs) are present.
Details
Evidence
joker tools: [Write]—matches stated purpose of delivering humor.
studio-coach tools: [Task, Write, Read]—aligns with agent coordination and coaching responsibilities without excessive scope.
MCP Risk
Low
No MCP servers are defined in the profile.
No external binary execution, network access, or filesystem operations are configured.
ForgeCat