managing-launchdarkly

Warn

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill defines shell scripts that interpolate positional parameters directly into command strings within the ld_api helper function.
  • Evidence: Variables like PROJECT_KEY and ENV_KEY are assigned from $1 and $2 and then passed to the ld_api function where they are used in a curl command string.
  • Risk: This interpolation allows for potential shell command injection if the input contains expansion characters like $(...) or backticks.
  • [EXTERNAL_DOWNLOADS]: The skill performs network requests to the LaunchDarkly API to retrieve and manage feature flags.
  • Evidence: The ld_api function uses curl to communicate with https://app.launchdarkly.com/api/v2.
  • Note: This involves a well-known service and is a core part of the skill's intended functionality.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by reading and displaying untrusted data from an external service.
  • Ingestion points: LaunchDarkly API responses fetched in discovery and analysis phases (SKILL.md).
  • Boundary markers: Absent; data from the API is processed by jq and printed directly to the output.
  • Capability inventory: The skill has the ability to execute shell commands via curl and jq.
  • Sanitization: No evidence of data sanitization or escaping before the ingested content is presented to the agent context.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 17, 2026, 01:28 AM
Security Audit — agent-trust-hub — managing-launchdarkly