Total Skills
2
Skills published by GreyFlames07 with real stars/downloads and source-aware metadata.
Total Skills
2
Total Stars
0
Total Downloads
0
Comparison chart based on real stars and downloads signals from source data.
forge-armour
0
forge-implement
0
Use this skill when the user wants to harden completed Forge specs after forge-audit and before implementation so the system meets strong security standards. Activates on phrases like "security review the specs", "harden the project", "run forge-armour", "apply security standards", "check the spec against best practice", or "prepare the spec for secure implementation". Runs a security posture review across project, module, and atom levels; builds a project security profile; maps risks to concrete spec changes; proposes project-level and module-level controls; and only writes approved changes after asking the human operator. Does NOT implement code and does NOT silently edit specs.
Use this skill when the user wants to translate a completed, audited Forge spec corpus into working implementation — code files, tests, and optionally git commits. Activates on phrases like "implement the project", "build the code", "let's implement", "run forge-implement", or after forge-audit completes cleanly with no open blocking findings. Recommends a `forge-armour` pass before implementation when the project has meaningful security exposure, but does not require it. Dispatches the forge-test-writer and forge-implementer skills via Task subagents (subagent_type: general-purpose; skill name invoked in the prompt) in parallel per the dependency graph, enforcing strict test-before-implementation isolation. Tests cover unit, integration, and system levels. Does NOT modify specs — specs are frozen at run start. Does NOT make implementation decisions — those live in the architecture section of the generated plan (human-confirmed) or in the specs themselves.