daily-papers-review
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands for git automation including
git add,git commit, andgit push. These actions are gated by specific configuration flags (GIT_COMMIT_ENABLED,GIT_PUSH_ENABLED) and use hardcoded or system-derived arguments rather than untrusted user input. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from
/tmp/daily_papers_enriched.jsoncontaining paper titles, summaries, and authors and interpolates it into prompts to generate reviews. - Ingestion points:
/tmp/daily_papers_enriched.json(Phase 5). - Boundary markers: Absent.
- Capability inventory: File write to Obsidian vault,
git add,git commit,git push(Phase 6). - Sanitization: Absent; the skill relies on the LLM to filter and summarize content based on the provided JSON fields.
Audit Metadata