Results for “Tekkiiiii/the-agency”

Each agent loop's own context %, measured per request; a one-line CONTEXT_PCT_ALERT rides that agent's next tool result when it crosses 70% or 80%

Band above the prompt: context %, tokens, model, cost, prompt-cache size/share/TTL countdown, 5h and weekly usage-window reset timers and next-request estimate; /ctx opens a detail pane; toast when a background subagent finishes; Jev router line (today's calls, tokens, USD, Haiku fallback) and outage warning, hidden unless the router usage log exists

Stall detector per agent loop: 5 identical tool calls in a row warn that loop and leave the session-state.json stall marker for the next SessionStart

Spawn log (spawn_start/launched/end + cost, native agentId lineage), artifact check and in-session RESPAWN_REQUEST pickup for every subagent, background included

Startup token diet: rewrites the long CAVEMAN / PONYTAIL MODE ACTIVE hook-context injections into ~150-token compact rules (level kept, pointer to /caveman or /ponytail)