competitive-market-shift-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill operates within defined operational boundaries, relying on platform-provided tools for data retrieval and adhering to a strict reasoning framework. The author (LMTY) is consistent with the vendor resource naming patterns described in the platform context.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted data including competitor announcements, news, and market signals.
- Ingestion points: Data enters the system via the
get_reportandget_report_changestools, as well as through user-provided files and web research specified inreferences/CONTEXT.mdandreferences/LMTY.md. - Boundary markers: The instructions in
SKILL.mdandreferences/TRUST.mdemphasize labeling inferences and restating hypotheses, which serves as a logical separator, though explicit character-level delimiters for external input are not defined. - Capability inventory: The skill's capabilities are limited to analytical reasoning and producing markdown artifacts. There is no evidence of arbitrary command execution, privilege escalation, or network exfiltration routines.
- Sanitization: The methodology relies on human-in-the-loop verification and explicit evidence quality checks (Step 2 and Step 5 in
SKILL.md) to mitigate misinformation, but does not implement automated sanitization of data strings. - Risk assessment: The structural risks associated with processing external market data are mitigated by the skill's restricted output format and the lack of high-privilege tool access.
Audit Metadata