The Agent Skills Directory

[SAFE]: The skill implements a legitimate workflow for context recovery, utilizing sub-agents to process data in an isolated context and prevent excessive data load in the primary session. No malicious patterns or unauthorized behaviors were identified.
[INDIRECT_PROMPT_INJECTION]: The skill processes content from previous session logs and external files, which represents an indirect prompt injection surface. This is a standard risk for context-reading tools and is mitigated here by its intended use-case.
Ingestion points: Parent session lineage files, issue specifications, work logs, and code files (identified in SKILL.md instructions).
Boundary markers: Absent; the instructions do not specify the use of delimiters when passing recovered context to the sub-agent or main agent.
Capability inventory: Uses the session-searcher sub-agent, aichat:session-search skill, and general file system access for markdown and code files.
Sanitization: Absent; the skill relies on the agent's internal interpretation and summarization of the retrieved text content.

recover-context