pilot-release-management-setup

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses the clawhub package manager to install several dependent skills, such as pilot-github-bridge and pilot-announce, which are required for the agent roles.
  • [COMMAND_EXECUTION]: Configuration is performed via Bash using the pilotctl utility to define hostnames, write configuration manifests, and establish trust handshakes between agents.
  • [DATA_EXFILTRATION]: The pipeline is designed to transmit release announcements to external services including Slack and webhooks via port 443.
  • [PROMPT_INJECTION]: The skill establishes an automated pipeline that ingests untrusted content from GitHub Pull Requests and commits, creating a surface for indirect prompt injection as this data is processed and broadcasted by subsequent agents.
  • Ingestion points: Pull Request metadata and commit history fetched by the pilot-github-bridge skill (referenced in SKILL.md).
  • Boundary markers: No delimiters or safety instructions are included in the manifest templates to isolate external content from agent instructions.
  • Capability inventory: The pipeline has the capability to publish data to external communication channels via pilot-slack-bridge and pilot-webhook-bridge (SKILL.md).
  • Sanitization: The setup procedure does not specify any sanitization or validation steps for the data processed from the source repository.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 07:29 PM
Security Audit — agent-trust-hub — pilot-release-management-setup