video-digest-segment
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to ingest and analyze transcript and schedule data from the local environment, which presents a surface for indirect prompt injection if the source text contains malicious instructions aimed at the agent.\n
- Ingestion points: The agent is instructed to read from
{ITEM_DIR}/transcript/timeline_30s.txt,{ITEM_DIR}/schedule.txt, and{ITEM_DIR}/transcript/transcript.tsvas specified inSKILL.mdandreferences/segment-prompt.md.\n - Boundary markers: The skill's instructions do not include explicit delimiters or safety prompts to treat the ingested transcript content as untrusted or to ignore instructions embedded within the data.\n
- Capability inventory: The skill directs the agent to perform file read operations on transcripts and write output files to
{ITEM_DIR}/talks.jsonand{ITEM_DIR}/sessions/boundaries.json.\n - Sanitization: No sanitization, filtering, or validation logic is provided to clean the transcript content before it is processed by the agent.\n- [NO_CODE]: The skill consists entirely of markdown documentation and instructions for an LLM agent; it does not include any Python scripts, JavaScript files, binaries, or other executable code.
Audit Metadata