The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill has a potential attack surface as it processes information from existing vault files to inform its logic and check for contradictions. 1. Ingestion points: Reads vault configuration and doctrine from 99_Meta/structure-doctrine.md and active decisions from 02_Command-Base/Decisions/. 2. Boundary markers: Includes explicit instructions for the agent to ask the user for answers rather than providing them autonomously. 3. Capability inventory: Limited to writing new markdown notes in 00_Inbox/ and 04_Methodology/, and appending log lines to 99_Meta/memory.md and 99_Meta/filing-log.md. 4. Sanitization: Mandates that the owner must see the exact words before the agent writes to the vault's memory or profile files.
[SAFE]: No patterns of obfuscation, data exfiltration, or unauthorized command execution were detected. The skill's operations are confined to a local organizational structure consistent with its stated purpose as a productivity tool.

breakthrough-session-report