The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests content from external files like spec.md and delta/ documents, which represents a potential surface for indirect prompt injection. * Ingestion points: Reads AGENTS.md, docs/glossary.md, docs/adr/, docs/pdr/, spec.md, seed.md, and delta/ files. * Boundary markers: No explicit markers are used to delimit external content, although the skill strictly defines the output plan structure. * Capability inventory: Limited to reading files and writing a single markdown file; no network or shell execution capabilities are available. * Sanitization: No specific sanitization or filtering of ingested data is described.
[SAFE]: All identified file operations and information gathering steps are consistent with the skill's primary purpose of implementation planning. No patterns of data exfiltration, command execution, or obfuscation were found.

plan-brief