The Agent Skills Directory

[COMMAND_EXECUTION]: The reference documentation references/decision-frontier-elicitation.md specifies the use of a local Python script scripts/validate_decision_frontier.py for validating decision graphs. This is a functional utility required for the skill's workflow.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted project documentation, creating a surface for indirect prompt injection. 1. Ingestion points: The skill reads stakeholder_register.md, vision.md, features.md, and glossary.md in Step 1. 2. Boundary markers: The instructions specify a rigid Output Format Specification using markdown headers and stable identifiers (EL-XXX) to structure and delimit data. 3. Capability inventory: The skill performs local file read/write operations and local script execution. 4. Sanitization: No explicit data sanitization is mentioned; the skill relies on structured elicitation protocols and the template structure to maintain integrity.
[SAFE]: No malicious patterns, obfuscation, or unauthorized exfiltration attempts were found during the analysis of the skill's core instructions and references.

02-elicitation-toolkit