srobroek
@srobroek
Public Skills
python-quality
by srobroek
Use to run Python format, lint, type-check, and test commands with the project toolchain.
rust-quality
by srobroek
Use to run Rust format, lint, and test checks with the project toolchain.
journey-init
by srobroek
"Use when setting up a user-journeys directory for a repo that has none: investigates, interviews, and scaffolds the format spec."
debate
by srobroek
Use when stress-testing an architectural decision, technology choice, or feature proposal from both sides before committing.
license-picker
by srobroek
Selects an OSI-approved license from project constraints and explains implementation tradeoffs. Use for new repositories, relicensing, or “what license should I use?”
brownfield-project
by srobroek
Retrofit an existing repository into APM-managed agentic tooling. Use when onboarding or repairing a brownfield repo that needs agents, skills, or hooks added.
omp-plugin-authoring
by srobroek
Use when creating, packaging, installing, or debugging an OMP plugin (skills, rules, agents, extensions) or a marketplace.
build-formula
by srobroek
Use when authoring a bd formula, choosing whether work should be a formula at all, or debugging one that pours the wrong DAG.
omp-surface-choice
by srobroek
Use when deciding whether guidance becomes a TTSR rule, a tool_call gate, a registered tool, a slash command, a skill, or stays prose.
journey-write
by srobroek
Use when authoring or amending user journeys from feature evidence, grilling on unknowns and checking the definition-of-ready.
eli5
by srobroek
Explain a topic at five depth levels from metaphor to frontier. Use when asked "ELI5", "explain X", "what is X", or "how does X work".
audit-steering
by srobroek
Audit and optimize agent-facing markdown for drift, duplication, and token waste. Use when asked to audit steering, review agent config, or optimize skills.
adr
by srobroek
Register an architecture decision as a beads decision bead. Use when a hard-to-reverse choice is made, when superseding one, or when bd lint reports findings.
verify
by srobroek
Run and report a final local verification pass. Use when asked to verify, test everything, check readiness, or prove changes safe to hand off.
journey-campaign
by srobroek
"Use when running journey or issue-driven validation at fleet scale: fan out validators, fix defects, and re-validate until green."
journey-verify-changed
by srobroek
Use when validating only the journeys affected by a diff, PR, or branch change rather than the full journey suite.
journey-consolidate
by srobroek
"Use when advancing a journey checkpoint with human approval: flush delta-log entries, prune old runs, and regenerate the index."