truefoundry-tracing

Pass

Audited by Gen Agent Trust Hub on Sep 23, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the traceloop-sdk (Python) and @traceloop/node-server-sdk (Node.js) packages from official registries. It also provides automated installation for the vendor's own CLI tool (truefoundry). These are standard operations for setting up observability in a development environment.- [COMMAND_EXECUTION]: The skill utilizes a helper script tfy-api.sh to interact with the TrueFoundry API. This script is designed with security in mind, using a custom parser for .env files instead of potentially dangerous shell sourcing, and includes validation to prevent path traversal attacks.- [INDIRECT_PROMPT_INJECTION]: The skill scans application configuration files (such as package.json or requirements.txt) to identify used libraries and then modifies entry point files to add instrumentation code. While this represents a data ingestion surface, the actions are restricted to detection and adding standard tracing initialization code.
  • Ingestion points: Project configuration files (requirements.txt, package.json, etc.) and application source code.
  • Boundary markers: Not applicable as the skill performs targeted detection of library names.
  • Capability inventory: Shell execution for package management and API calls; file system write for code instrumentation.
  • Sanitization: The skill performs specific string matching for known library names rather than executing content from the scanned files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 23, 2026, 03:58 PM