fittings-data-column-profile

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFECREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill documentation notes that the FITTINGS_PRIVATE_KEY, used for the x402 payment protocol, sits in plaintext within the environment. This practices poses a risk of credential exposure if the execution environment is accessed by unauthorized processes or if instructions leak environment variables.\n- [DATA_EXFILTRATION]: The skill is designed to send local data (CSV or NDJSON format) to a remote endpoint at https://fittings.sh/v1/data/column-profile for processing. While this is the core utility of the skill, it constitutes external transmission of potentially sensitive dataset contents.\n- [EXTERNAL_DOWNLOADS]: The skill references external Node.js packages @x402/core and @x402/evm as dependencies for the payment handshake. It also references a remote script https://fittings.sh/skill/call.mjs for documentation purposes.\n- [PROMPT_INJECTION]: The skill processes untrusted external data via the data parameter. This creates an indirect prompt injection surface.\n
  • Ingestion points: CSV or NDJSON string input provided to the POST /v1/data/column-profile call.\n
  • Boundary markers: None identified in the provided instructions to delimit data from instructions or warn the agent about embedded content.\n
  • Capability inventory: The skill uses curl to transmit data and receive profiling results.\n
  • Sanitization: No specific sanitization or validation of the input data is described before it is sent to the remote service.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 10:32 PM