palantir-webhooks-events

Pass

Audited by Gen Agent Trust Hub on Sep 12, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data from Palantir Foundry webhooks, which creates a potential surface for indirect prompt injection if the source data is compromised.
  • Ingestion points: The handle_foundry_webhook function in the Flask endpoint ingests JSON payloads from external HTTP POST requests.
  • Boundary markers: The implementation correctly uses X-Foundry-Signature verification with HMAC-SHA256 and hmac.compare_digest, along with timestamp-based replay protection to verify the origin of data.
  • Capability inventory: The skill defines capabilities for Read, Write, Edit, and Bash (specifically curl). Python snippets utilize environment variables and network connections to Redis for idempotency.
  • Sanitization: While the skill authenticates the source, it does not explicitly sanitize the internal fields of the JSON payload before they are processed or logged, which is a standard surface for this category.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 12, 2026, 09:36 PM
Security Audit — agent-trust-hub — palantir-webhooks-events