flare
Warn
Audited by Gen Agent Trust Hub on Feb 22, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- EXTERNAL_DOWNLOADS (MEDIUM): The skill instructs users to install the
spatie/flare-clipackage via Composer. While Spatie is a known PHP developer, it is not included in the predefined list of trusted organizations. - COMMAND_EXECUTION (SAFE): The skill executes various
flarecommands to interact with the Flare API. These actions are aligned with the skill's stated purpose of managing error tracking and monitoring. - PROMPT_INJECTION (LOW): The skill is vulnerable to Indirect Prompt Injection (Category 8). It ingests potentially untrusted data from the Flare API, such as exception messages and stack traces, which could be influenced by an external attacker to manipulate the agent's behavior. 1. Ingestion points:
flare list-project-errorsandflare get-error-occurrence. 2. Boundary markers: Absent. 3. Capability inventory: Subprocess execution of theflareCLI tool. 4. Sanitization: No sanitization of the error data or traces is performed before processing.
Audit Metadata