octocode-awareness
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a robust repository awareness system using a local SQLite database located at
~/.octocode/memory/awareness.sqlite3to track tasks, file presence, and coordination signals. - [COMMAND_EXECUTION]: The skill utilizes a suite of bundled shell scripts (e.g.,
scripts/hooks/pre-edit.sh) and Node.js utilities (hook-runner.mjs,awareness.mjs) to automate lifecycle events. These executions are scoped to the skill's own logic and the vendor's CLI tools. - [EXTERNAL_DOWNLOADS]: Instructions guide the user to install the
@octocodeai/octocode-awarenessand@octocodeai/octocode-skillspackages via NPM. These are official vendor resources necessary for the CLI and research capabilities of the skill. - [PROMPT_INJECTION]: No malicious prompt injection patterns, safety bypasses, or instruction overrides were detected. The skill uses structured instructional language to guide agent behavior during repository work.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect injection by reading workspace files and tool outputs. However, it mitigates this risk by emphasizing 'compact' receipts, using boundary markers, and focusing on deterministic state management.
Audit Metadata