clawteam-directors-orchestrator

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The utility script scripts/match_directors.sh accepts user input through the first command-line argument and processes it using a shell pipeline (echo "$DEMAND" | grep). This pattern is vulnerable to shell command injection, as the shell will evaluate command separators (like ;, &, or |) or subshell syntax (backticks or $()) within the input string, potentially allowing for arbitrary code execution in the environment where the script is run.- [PROMPT_INJECTION]: The skill architecture is susceptible to indirect prompt injection due to its handling of external data.
  • Ingestion points: User demand strings are processed directly by the match_directors.sh script and interpolated into the orchestration logic defined in SKILL.md.
  • Boundary markers: No delimiters or explicit instructions to ignore embedded commands are used to encapsulate the user-provided demand string.
  • Capability inventory: The skill possesses the capability to execute local shell scripts and write files to the local filesystem (as seen in generate_director_skills.sh).
  • Sanitization: No sanitization, escaping, or filtering is performed on the user input before it is utilized in shell commands or prompt instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 03:45 PM
Security Audit — agent-trust-hub — clawteam-directors-orchestrator