continuous-learning-v2

Pass

Audited by Gen Agent Trust Hub on Oct 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPERSISTENCE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill captures session history (prompts, tool inputs and outputs) into local observation logs and uses a background sub-agent (Claude Haiku) to synthesize this data into behavioral rules ("instincts"). This creates a vulnerability surface where malicious content from external tool outputs could be captured and processed as instructions for future behavior.
  • Ingestion points: Observations are collected in ~/.claude/homunculus/projects/<hash>/observations.jsonl (see hooks/observe.sh).
  • Boundary markers: The background agent is provided with explicit instructions and delimiters in its prompt (see agents/observer-loop.sh), though it ultimately processes raw session data.
  • Capability inventory: The sub-agent is granted permission to use the Read and Write tools to create or update instinct files in the homunculus directory.
  • Sanitization: The hooks/observe.sh script implements a regex-based secret scrubber (_SECRET_RE) that redacts common API key, token, and password patterns before they are persisted to the observation logs.
  • [EXTERNAL_DOWNLOADS]: The instinct-cli.py utility includes an import command that fetches instinct definitions from remote URLs using urllib.request.urlopen. While this is a user-triggered feature, it allows for the ingestion of behavioral rules from arbitrary external sources.
  • [COMMAND_EXECUTION]: The skill executes system commands such as git for project context detection and the claude CLI to perform the background observation and analysis tasks.
  • [PERSISTENCE]: The skill implements a persistent background service model using nohup and PID files (.observer.pid) to maintain a continuous learning loop across different terminal sessions.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 14, 2026, 03:03 PM