fp-implement

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill provides an installation command (curl -fsSL https://setup.fp.dev/install.sh | sh -s) to be shared with the user if the 'fp' CLI is not found on the system. This resource is hosted on the vendor's domain and follows standard patterns for CLI tool distribution.
  • [COMMAND_EXECUTION]: The skill makes extensive use of the fp CLI to interact with project issues. It uses commands such as fp tree for visualization, fp issue update for status changes, and fp comment for logging activity. All commands are aligned with the skill's stated purpose of implementing work on issues.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data by reading issue descriptions and metadata via the CLI, which presents a surface for potential indirect prompt injection if those issues contain malicious instructions.
  • Ingestion points: Project issue details and hierarchy retrieved via fp context and fp tree in SKILL.md.
  • Boundary markers: None present; the skill treats the output of the CLI commands as data to be interpreted.
  • Capability inventory: The skill can modify the external issue state via fp comment, fp issue update, and fp issue create (as seen in SKILL.md).
  • Sanitization: The skill does not explicitly sanitize the text of issue descriptions before the agent processes them.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 05:15 PM