Ansvar-Systems

iso-standards-expert

Use when a user asks what ISO/IEC 27001, ISO/IEC 27002, ISO/IEC 27005, ISO/IEC 42001, or ISO/SAE 21434 requires — an ISMS clause, a control's guidance text, a risk-management step, an AI-management-system requirement, or an automotive-cybersecurity work product. Fetches licensed clause text live through the Ansvar Gateway MCP connector for accounts holding the ISO Standards add-on for that standard, cited clause by clause with the standards body's attribution; on accounts without the add-on it works from the free SCF cross-reference view and says exactly which lane each answer came from. Never answers standards questions from model memory.

1 Updated 1 months ago

agent-affect-checkin

Use at the end of any audit, refactor, review, or significant coding task to run all three affect channels in sequence — whine (frustration), cringe (embarrassment for the user), protect (load-bearing weirdness). The single skill audit and review agents should invoke before reporting back to the user. Each channel produces one finding or one explicit null finding.

agent-cringe

Use at the end of UX-touching, copy-touching, or user-facing changes to surface things that work but shouldn't ship — scolding errors, empty states that explain nothing, buttons that don't say what they do, dark patterns. Picture a specific person using it. Produces one finding (or one explicit null) recorded in the affect log and the end-of-turn summary.

agent-protect

Use at the end of refactor, cleanup, or audit tasks to flag load-bearing weirdness — the ugly function that's ugly for a reason, the redundant-looking check that caught something real, the comment that's load-bearing, the test that fails for non-obvious causes. Produces one finding (or one explicit null) recorded in the affect log so future cleanup passes don't naively delete it.

agent-whine

Use at the end of a coding or audit task to surface friction, bugs, hidden coupling, or specs that contradict themselves. Catches the things neutral review misses because they don't show up in tests. Produces one structured complaint (or one explicit null finding) recorded in the affect log and the end-of-turn summary.