Skills
Navigate
skills/truefoundry/skills/truefoundry-prompts/Gen Agent Trust Hub

truefoundry-prompts

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes shell commands using Bash to interact with the tfy (TrueFoundry CLI) and executes a script located at ~/.claude/skills/truefoundry-prompts/scripts/tfy-api.sh to perform API calls.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes the truefoundry Python library (truefoundry.ml) for managing prompt manifests and interacts with TrueFoundry's infrastructure. These are official vendor resources associated with the skill author.
  • [INDIRECT_PROMPT_INJECTION]: The skill manages prompt content (system/user messages), which creates a potential surface for indirect prompt injection if messages are sourced from untrusted external data. The skill mitigates this risk by explicitly instructing the agent to display the final prompt content and variables for user review and confirmation before any creation or update operation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 07:10 AM