gifgrep
Warn
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the agent to install the
gifgrepbinary from external, non-trusted sources: a Homebrew tap (steipete/tap/gifgrep) and a Go module (github.com/steipete/gifgrep/cmd/gifgrep@latest). - [COMMAND_EXECUTION]: The skill executes the
gifgrepcommand to search for GIFs, interact with a TUI, and process images into stills or contact sheets. This involves running external binaries with user-controlled arguments. - [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection from data retrieved via GIF provider APIs.
- Ingestion points: Metadata such as titles and tags are fetched from Tenor and Giphy APIs (SKILL.md).
- Boundary markers: No markers are defined to separate untrusted API data from the agent's instructions.
- Capability inventory: The skill allows execution of the
gifgrepbinary and file system access to~/Downloads(SKILL.md). - Sanitization: No sanitization is performed on the data retrieved from external GIF providers before processing or display.
Audit Metadata