upstream-patches

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute git commands and a local script ./scripts/upstream.sh to manage the patch stack and repository state.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external patch files which could contain hidden instructions targeting the agent.
  • Ingestion points: The agent reads patches/*.patch using rg, grep, and sed to find and inspect patches.
  • Boundary markers: There are no instructions to the agent to treat patch content as untrusted or to use delimiters.
  • Capability inventory: The agent is granted capabilities to delete files (rm), modify git history, and run a local management script.
  • Sanitization: No sanitization or validation of the ingested patch content is performed.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 07:56 AM
Security Audit — agent-trust-hub — upstream-patches