skills/jssblck/agents/using-sops/Gen Agent Trust Hub

using-sops

Pass

Audited by Gen Agent Trust Hub on Sep 24, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to download the sops and age binaries directly from their official GitHub release pages.
  • Evidence: Downloads sops from github.com/getsops/sops and age from github.com/FiloSottile/age in references/setup.md.
  • [REMOTE_CODE_EXECUTION]: The setup documentation describes installing downloaded binaries and modifying their execution permissions.
  • Evidence: curl -fsSL -o ~/.local/bin/sops ... && chmod +x ~/.local/bin/sops and curl ... | tar xz in references/setup.md.
  • [COMMAND_EXECUTION]: The skill provides a wrapper command (pnpm secrets exec) that spawns child processes with decrypted secrets injected into their environment.
  • Evidence: Workflow describes executing commands like node apps/worker/src/main.ts with decrypted variables via the exec utility.
  • [CREDENTIALS_UNSAFE]: The skill is designed to manage high-value credentials, specifically age private keys (AGE-SECRET-KEY-1...) and encrypted .env files. It interacts with sensitive paths such as ~/.config/sops/age/keys.txt.
  • Evidence: Instructions for storing private keys in the local config directory and using them for decryption in SKILL.md and references/setup.md.
  • [PRIVILEGE_ESCALATION]: The documentation includes instructions for using sudo to perform administrative tasks, such as creating system directories and using systemd-creds for secure credential storage.
  • Evidence: sudo mkdir -p /etc/credstore.encrypted and sudo systemd-creds encrypt in references/setup.md.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted data (encrypted secrets) which could potentially contain malicious instructions if the source repository or its secrets are compromised.
  • Ingestion points: Reads data from secrets/dev.env and secrets/prod.env in SKILL.md.
  • Boundary markers: Relies on SOPS integrity checks (MAC) and explicit command wrappers.
  • Capability inventory: Capability to execute child processes via pnpm secrets exec and write to local files via pnpm secrets set.
  • Sanitization: The wrapper tools/secrets.ts is described as stripping SOPS_AGE_KEY from the child environment to prevent credential leakage.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 24, 2026, 01:57 PM