teamai-share-learnings

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the teamai contribute command-line utility to push session summaries to a remote team repository. This behavior is consistent with the skill's stated purpose of knowledge sharing and relies on internal vendor tooling.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes session history and tool outputs to generate summaries. While this constitutes an attack surface for indirect prompt injection, the risk is minimal as the output is constrained by a specific Markdown template and intended for human-reviewed knowledge bases.
  • Ingestion points: Session history and tool usage data (from SKILL.md instructions) are used to generate the summary content.
  • Boundary markers: No explicit delimiters are used to separate session data from the summary instructions.
  • Capability inventory: The skill has the capability to execute shell commands via the teamai CLI and write to temporary files.
  • Sanitization: There is no explicit sanitization shown for the content generated from session data before it is passed to the command-line tool's title argument.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 01:06 PM
Security Audit — agent-trust-hub — teamai-share-learnings