New subsection in Discussion framing Wiggum Breaks as the formal boundary
between autonomous and human-supervised operation. Derives HITL from
convergence theory rather than pre-defined approval gates. Covers
oscillation, divergence, and repeated shadow detection as provably
unproductive conditions that trigger human escalation.
Replaces generic "circuit breaker" with "Wiggum Break" — policy enforcement
halt condition named after Chief Wiggum (policy + Ralph Loop's dad).
Hard breaks (immediate halt) and soft breaks (finish then halt) with
wiggum.break event type. Updated both papers and shadow-detection skill.
Incorporate findings from literature survey: Six Sigma Agent (arXiv:2601.22290)
as the only prior explicit PM/OM-named framework, AgileCoder for Scrum sprints,
Reflexion as implicit PDCA, CAMEL for role theory.
LaTeX paper describing the archetypal role system, PDCA quality cycles,
shadow detection framework, attention filters, convergence detection,
and effectiveness scoring. References Lu et al. 2026 (Assistant Axis)
for persona stability grounding.
- archeflow-decision.sh records decision points during runs
- archeflow-replay.sh: timeline, whatif, compare commands
- What-if replay with adjustable archetype weights
- /af-replay skill for interactive use
- Tests in archeflow-replay.bats
- archeflow-decision.sh: convenience wrapper for logging PDCA decision points
- archeflow-replay.sh: timeline view and weighted what-if replay for recorded runs
- archeflow-event.sh: add decision.point usage example
- archeflow-dag.sh: render decision.point events in DAG output
- Extend shadow-detection with 3-layer corrective action framework:
archetype shadows, system shadows (tunnel vision, echo chamber, etc.),
and policy boundaries (checkpoints, budget gates, circuit breakers)
- Rewrite CLAUDE.md with proper guardrails (DO/DO NOT, skill writing rules,
200-line max per skill, no bash pseudo-code in skills)
- Update plugin.json to v0.8.0 with consolidated 19-skill list
- Update README architecture tree and skills reference
- Update using-archeflow version string to v0.8.0 / 19 skills
- Remove 8 empty skill directories (absorbed into run skill)
Remove duplicated routing tables, verbose JSON event examples,
writing/prose domain template (belongs in domains/colette-bridge),
--start-from section (belongs in run skill), and redundant checklist.
Consolidate three Agent() templates into one compact template.
Preserve all routing rules, decision logic, and feedback format.
Consolidate the attention-filters skill (122 lines) into check-phase,
reducing check-phase from 234 to 110 lines. Removed verbose bash code
blocks, 30-line consolidated output example, re-check protocol (belongs
in act-phase), and motivational section. Updated all references in
README, plugin.json, using-archeflow, and colette-bridge.
- README: lead with af-sprint (parallel multi-project), af-review (post-impl quality)
- Sprint skill: L/XL code tasks use feature-dev style (explore→plan→impl→self-review)
instead of PDCA. Reserve PDCA for writing/research domains.
- Session start: route to af-sprint/af-review/af-run based on task type
- Explicitly state: for single-feature dev, use feature-dev plugin instead
Standalone bash script that extracts git diffs for af-review without
PDCA orchestration. Supports --branch, --commit, and uncommitted modes.
Reports stats (files/lines changed) to stderr, diff to stdout.
- Auto-select: fast workflow now maps to pipeline strategy (was falling through to pdca)
- Evidence validation: check for missing evidence markers, not just banned phrases
- Remove sed-based artifact mutation (avoids table row corruption), track downgrades in events only
- Pipeline verify: explicit merge guard prevents merging before tests/re-review pass
