Block a user
feat: telos-load skill — harness-agnostic TELOS session context loading
feat: regulatory-risk-assessment skill — CAD compliance gate
fix: ntfy iOS broken — basic auth setup + topic taxonomy + agent-approval action buttons
Koala execution prompt
You are starting a focused infrastructure session on koala.
First: brain_context wing=telos limit=8
## Session objective
Fix ntfy so the iOS app works and Mathias…
feat: regulatory-risk-assessment skill — CAD compliance gate
Koala execution prompt
You are starting a focused skills session on koala.
First: brain_context wing=telos limit=8
## Session objective
Add two new skills to mathias/skills, commit, push,…
Should SKILL.md define explicit escalation / failure conditions?
ADR: centralize Google OAuth via Authentik social source — deferred until third Google service
feat: email-triage-agent — multi-account Gmail digest + aging-based auto-archive via LLM classification
fix: ntfy iOS broken — basic auth setup + topic taxonomy + agent-approval action buttons
[handover] Brain-MCP intent↔interface study — agent column + merge scaffold ready for review
[handover] Brain-MCP intent↔interface study — agent column + merge scaffold ready for review
Human column delivered — merge complete
human-intent-column.md produced from claude.ai session history analysis (15 knowledge-acts across 6 sessions, 2026-06-15/16). Stored in brain:…
[handover] Brain-MCP intent↔interface study — agent column + merge scaffold ready for review
LANGUAGE.md: fix LiteLLM gateway never-say (stale endpoints, not piguard)
LANGUAGE.md pilot: hand-written vocabulary + caveman rubric (Phase 1)
Follow-on fix opened as PR #21 (mathias/tapir#21) — do not merge without review.
This PR (#19) was already merged, so the LiteLLM never-say correction lands on a…