st-generate-tasks
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill executes a script (
scripts/check-for-updates.cjs) that performs an HTTP GET request to the GitHub API (api.github.com/repos/e0ipso/strikethroo/releases/latest) to check for newer versions of the tool. This is a standard update-check mechanism targeting the developer's official repository. - [INDIRECT_PROMPT_INJECTION]: The skill decomposes project plans into tasks by reading and analyzing user-provided markdown files in
Step 4. This creates an attack surface where malicious instructions embedded in a project plan could potentially influence the agent's behavior during the task generation process. - Ingestion points: Project plan files and configuration files (e.g.,
STRIKETHROO.md) are read into the agent's context inStep 3. - Boundary markers: The skill uses an 'anti-rationalization' table to enforce specific logic, but lacks strong delimiters to distinguish between data and embedded instructions in the input plans.
- Capability inventory: The agent can write new task files to the file system in
Step 10and execute various helper scripts in thescripts/directory. - Sanitization: No explicit sanitization or filtering of the input plan content is performed before processing.
Audit Metadata