The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructions provide a command to download and execute a shell script from https://edgespeak.com/install.sh using curl -fsSL ... | sh. This is an intended distribution mechanism for the tool and the domain is associated with the skill's author.
[EXTERNAL_DOWNLOADS]: The skill fetches installation assets, license verification metadata, and machine learning models such as lattice-1-text-segmenter from vendor-controlled servers.
[COMMAND_EXECUTION]: The skill interacts with the local operating system by executing edgespeak-cli for operations like login, activation, and text segmentation, occasionally communicating with a local gateway at 127.0.0.1:1117.
[PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection. 1. Ingestion points: text blocks and transcript JSON files provided via --text, --file, and --transcript flags. 2. Boundary markers: Absent; there are no instructions to the model to ignore content within these inputs. 3. Capability inventory: The agent has the ability to execute shell commands and write files to the local disk. 4. Sanitization: No data sanitization or input validation steps are documented in the skill instructions.

edgespeak-segment