The Agent Skills Directory

[SAFE]: The skill serves as a documentation and organizational standard for roadmap management. It instructs the agent on file placement (docs/ directory) and naming conventions without introducing executable scripts or network operations.- [NO_CODE]: The skill consists entirely of instructional markdown and schema definitions for a JSON state file. It does not include or reference any external scripts, binaries, or package dependencies.- [INDIRECT_PROMPT_INJECTION]: The skill defines structures for documents (PRDs, roadmaps) that may eventually contain external user input.
Ingestion points: Feature-level and step-level markdown files in the docs/ directory.
Boundary markers: Not explicitly defined in this convention skill.
Capability inventory: File creation, updates, and deletion restricted to documentation folders.
Sanitization: Not addressed within these conventions.
Note: As a metadata/convention skill, it does not process untrusted data directly, making the attack surface negligible.- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill mentions common sensitive patterns such as Jira ticket IDs (e.g., SCB-NNN) as part of ticket tracking, but it does not encourage the storage of credentials, secrets, or access to sensitive system paths. The defined storage location (docs/<feature-slug>/) is standard for project documentation.

roadmap-conventions