Integrate threat scanning for all memory content writes. Before persisting any memory content (add/replace/remove), scan the content for potentially dangerous patterns (injection attempts, sensitive data leakage, prompt injection in stored memory).
Acceptance Criteria
A ScanContent(content string) function or hook is called before every memory write
Scans cover: prompt injection patterns, sensitive data patterns (API keys, passwords, tokens)
Configurable to reject (block the write) or warn (allow with log) on suspicious content
Integration point at the MemoryManager level so it applies to all providers
Results are logged for audit
Config flag to disable scanning for development
Imported from beads archie-core-1786637494880-194-86baeb74 · GitHub #215 · type task · priority P2 · created 2026-08-08 · closed 2026-08-08
Integrate threat scanning for all memory content writes. Before persisting any memory content (add/replace/remove), scan the content for potentially dangerous patterns (injection attempts, sensitive data leakage, prompt injection in stored memory).
### Acceptance Criteria
A ScanContent(content string) function or hook is called before every memory write
Scans cover: prompt injection patterns, sensitive data patterns (API keys, passwords, tokens)
Configurable to reject (block the write) or warn (allow with log) on suspicious content
Integration point at the MemoryManager level so it applies to all providers
Results are logged for audit
Config flag to disable scanning for development
---
_Imported from beads `archie-core-1786637494880-194-86baeb74` · GitHub #215 · type task · priority P2 · created 2026-08-08 · closed 2026-08-08_
sam
added the mediumtask labels 2026-10-27 13:37:28 +00:00
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Integrate threat scanning for all memory content writes. Before persisting any memory content (add/replace/remove), scan the content for potentially dangerous patterns (injection attempts, sensitive data leakage, prompt injection in stored memory).
Acceptance Criteria
A ScanContent(content string) function or hook is called before every memory write
Scans cover: prompt injection patterns, sensitive data patterns (API keys, passwords, tokens)
Configurable to reject (block the write) or warn (allow with log) on suspicious content
Integration point at the MemoryManager level so it applies to all providers
Results are logged for audit
Config flag to disable scanning for development
Imported from beads
archie-core-1786637494880-194-86baeb74· GitHub #215 · type task · priority P2 · created 2026-08-08 · closed 2026-08-08