tally
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on shell commands using
npxto interact with the ClawLink CLI and Tally service for managing forms and webhooks.- [EXTERNAL_DOWNLOADS]: The skill's setup and usage instructions usenpx, which automatically fetches the@useclawlink/clipackage from the NPM registry at runtime.- [REMOTE_CODE_EXECUTION]: The execution ofnpx @useclawlink/clidownloads and runs remote code within the agent's environment, creating a dependency on the integrity of the NPM package.- [INDIRECT_PROMPT_INJECTION]: The skill defines tools that process external data from Tally, such as form responses, which could contain instructions designed to manipulate the agent's behavior. - Ingestion points:
tally_get_form_responses,tally_get_form_details, andtally_list_form_questionstools as described inSKILL.md. - Boundary markers: Absent; no specific instructions or delimiters are provided to the agent to ignore instructions embedded in the form data.
- Capability inventory: Shell command execution via
npx(documented inSKILL.md) allows the agent to perform various write and read operations based on processed data. - Sanitization: Absent; there is no mention of filtering or validating the content retrieved from Tally.
Audit Metadata