Replace the generic more-goals line with nonzero checkbox counts for hidden goals. Keep active and cancelled goals distinct and preserve the visible-goal limit.
Co-Authored-By: Pi/OpenAI <288921227+claudypoo@users.noreply.github.com>
Correct the overbroad instruction to replace implementation tasks with user outcomes. IKIWISI clarifies acceptance through examples; it must not turn Rust conversion, embeddings, or functioning search into vague readiness. Preserve execution approval through evidence-only edits.
Co-Authored-By: Pi/OpenAI <288921227+claudypoo@users.noreply.github.com>
Use "I know it when I see it" in initial and recurring planning instructions. Keep judging guidance under the goal and reject invented numerical gates, without discarding grounded numerical requirements.
Co-Authored-By: Pi/OpenAI <288921227+claudypoo@users.noreply.github.com>
Render a UI-only copy with Ctrl+O expansion and keep original user prompts unchanged. Verify exact prompt delivery and Ready role transition through real Pi RPC, plus an isolated Herdr display trial.
Co-Authored-By: Pi/OpenAI <288921227+claudypoo@users.noreply.github.com>
Unify Log boundaries, serialize completion edits, preserve immutable session state and solo recovery, and match stock worker execution events. Remove unused plan views and brittle prose checks; correct installation and context guidance.
Co-Authored-By: Pi/OpenAI <288921227+claudypoo@users.noreply.github.com>
Keep the hourly schedule unchanged. Advance the local nudge cycle only on prompt delivery and reset the unchanged-turn counter so upkeep repeats. Test compaction supersession, cadence, rotation and solo behavior.
Co-Authored-By: Pi/OpenAI <288921227+claudypoo@users.noreply.github.com>
Field calibration from three supervisor reports: the useful plan-change catch was a worker-ticked task the short-view hash no longer surfaced, while identity-line edits caused duplicate empty reviews. The notify digest keeps goals, tasks and evidence above the Log and drops worker identity lines. Reopening a signed goal still clears its sign-off.
Co-Authored-By: Pi/OpenAI <288921227+claudypoo@users.noreply.github.com>
Both panes now render the live widget glyphs, and the supervisor message is a real intercom steer in the user style instead of self-report prose.
Co-Authored-By: Pi/OpenAI <288921227+claudypoo@users.noreply.github.com>
A read-only reviewer launch no longer overwrites the implementation worker (reported by maniworker session 01a0809b): the first launch binds, resume of the same session refreshes it, later launches land in state.helpers and show in /goals status. launchPending is now a counter so concurrent launches keep takeover menus honest. Old persisted state migrates with an empty helper list.
Co-Authored-By: Pi/OpenAI <288921227+claudypoo@users.noreply.github.com>
Preserve wassname User ask and screenshot layout; retain bundled install instructions. 71 current tests pass, lint/typecheck clean. Exclude local backups/worktrees from test discovery.
Co-Authored-By: Pi/OpenAI <288921227+claudypoo@users.noreply.github.com>
Replace historical docs, trial output and dead runtime with current source, agent definition and reusable scripts. Bundle pinned worker, Intercom and scheduler dependencies via Pi manifest. Retain 71 current tests, including real RPC proposal/editor/Ready checks and evidence edge cases; lint/typecheck pass. Historical material remains available at f22d83c.
Co-Authored-By: Pi/OpenAI <288921227+claudypoo@users.noreply.github.com>
Preserve tasks in plan files; update README model illustration and prompt link. 176 tests pass, typecheck and lint clean.
Co-Authored-By: Pi/OpenAI <288921227+claudypoo@users.noreply.github.com>