The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes a local command-line utility bv to process data. This is a primary function of the skill and used for legitimate analysis tasks such as execution planning and graph metrics calculation.- [DATA_EXFILTRATION]: The skill accesses project-specific data in the user's documents folder. No network activity or patterns of data exfiltration were detected; all operations described are local to the host machine.- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it processes data from external files (.beads/beads.jsonl).
Ingestion points: Files located at ~/Documents/LLM CONTEXT/**/.beads/beads.jsonl processed by the bv tool.
Boundary markers: Not present; the instructions do not specify delimiters for separating tool output from agent instructions.
Capability inventory: The agent executes shell commands (bv) and makes prioritization recommendations based on the resulting data.
Sanitization: The skill does not describe any validation or filtering of the content within the issue tracker files before processing.

beads-viewer