hilos-share
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes session history and code changes to generate summaries, which presents a surface for processing untrusted data.
- Ingestion points: The skill reads the current session history and git diffs to generate content for sharing.
- Boundary markers: No explicit boundary markers or delimiters are used to wrap the ingested content.
- Capability inventory: The skill utilizes
list_channels,post_message, andgit diffcommands. - Sanitization: The skill contains explicit safety instructions to summarize content in 3-6 lines and strictly forbids sharing transcripts, tokens, or
.envfiles. - [DATA_EXFILTRATION]: The skill sends session data to the external Hilos service (
hilos.sh) via thepost_messagetool. - The transmission is the primary purpose of the skill and is triggered by user request.
- The instructions explicitly prohibit the exfiltration of sensitive credentials or environment files.
- [COMMAND_EXECUTION]: The skill invokes
git diff --statto retrieve information about project changes. - This is a standard development operation used here to inform the summary generation process.
Audit Metadata