The Agent Skills Directory

[SAFE]: The skill's behavior is consistent with its stated purpose of lead prioritization and evidence gathering. No suspicious instructions or malicious intent were found in the markdown body.
[NO_CODE]: This skill consists solely of natural language instructions in a markdown file and does not include any executable scripts, binaries, or configuration files that trigger code execution.
[DATA_EXFILTRATION]: The skill instructions involve resolving and enriching lead contact information (emails, phone numbers). This is the primary intended function of the skill and occurs within the agent's data processing context without evidence of unauthorized exfiltration to external domains.
[PROMPT_INJECTION]: The skill processes untrusted external data in the form of lead lists (Step 2), which creates a surface for indirect prompt injection. However, the instructions focus on structured data resolution which mitigates typical injection risks.
Ingestion points: Leads list containing prospect IDs, emails, or LinkedIn URLs (Step 2 in SKILL.md).
Boundary markers: No explicit delimiters or warnings to ignore embedded instructions are present.
Capability inventory: The workflow implies the use of person resolution, company resolution, and firmographic enrichment tools (Steps 3, 4, and 5).
Sanitization: No explicit sanitization or validation of the lead input is described in the instructions.

score-leads