The Agent Skills Directory

[PROMPT_INJECTION]: The skill ingests external Gong transcript data, creating a surface for indirect prompt injection.
Ingestion points: Call transcripts and metadata are fetched via the gong-calls and account-deep-dive tools as specified in SKILL.md.
Boundary markers: The instructions do not define specific delimiters for separating ingested transcript content from the agent's instructions.
Capability inventory: The skill utilizes the run-code, save-analysis, and provider-api-request capabilities.
Sanitization: Instructions specifically mandate filtering monologues to isolate customer speech and using compact excerpts to minimize the processing of raw external data.
[COMMAND_EXECUTION]: The skill directs the agent to use the run-code tool for data processing tasks, including joining datasets, deduplicating information, and formatting evidence. This usage is restricted to internal data manipulation consistent with the skill's purpose.

gong