specstory-session-summary
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs local file operations restricted to the
.specstory/history/directory. It uses standard utilities likels,grep, andtailto analyze session logs. There is no evidence of network exfiltration, credential harvesting, or unauthorized privilege escalation. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local history files which could contain untrusted instructions from previous sessions. However, the risk is negligible as the skill uses structured parsing logic to extract specific fields (Goal, Outcome, Files) for summarization rather than executing the content. Ingestion points: SpecStory history markdown files in
.specstory/history/*.md. Boundary markers: Uses_**User**_and---delimiters to scope content extraction. Capability inventory: Local file reading and shell-based string processing viagrepandtailtools. Sanitization: Relies on specific parsing patterns defined in the references to isolate data fields.
Audit Metadata