130c04fa58
feat: corrective action framework + CLAUDE.md rewrite + v0.8.0 cleanup
...
- 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)
2026-04-06 20:52:27 +02:00
752177528f
refactor: trim act-phase skill from 371 to 140 lines
...
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.
2026-04-06 20:50:59 +02:00
a1667633ad
Merge branch 'refactor/consolidate-check-phase-v2' into refactor/trim-secondary-skills
...
# Conflicts:
# skills/colette-bridge/SKILL.md
# skills/using-archeflow/SKILL.md
2026-04-06 20:50:31 +02:00
d94688ca1b
refactor: trim 11 secondary ArcheFlow skills from 3340 to 952 lines
...
Remove verbose YAML examples, bash pseudo-code, tutorial prose, and
motivational content from configuration/integration skills while
preserving all operational protocols, reference tables, and rules.
Skills trimmed: domains, colette-bridge, multi-project, cost-tracking,
git-integration, custom-archetypes, workflow-design, templates,
autonomous-mode, progress, presence.
2026-04-06 20:48:50 +02:00
c8bd55d97c
refactor: consolidate run skill — merge 8 skills into one self-contained PDCA orchestrator
...
Merge run + orchestration + plan-phase + do-phase + artifact-routing + process-log +
attention-filters + convergence + effectiveness into a single 459-line run/SKILL.md.
Before: run skill (890 lines) + 3 prerequisites (~1,300 lines) = ~2,200 lines of context.
After: one self-contained skill (459 lines) with zero prerequisites.
Preserved: PDCA flow, workflow selection, adaptation rules A1-A3, agent prompts,
attention filters, feedback routing, convergence detection, effectiveness scoring,
shadow monitoring, pipeline strategy, event reference, artifact naming.
Removed: verbose bash code blocks, shell variable tracking, resolve_model() function,
lib validation loops, evidence validation bash, redundant event emission blocks.
2026-04-06 20:44:46 +02:00
55de51aabe
chore: trim sprint and using-archeflow skills for context efficiency
...
sprint: 302 -> 164 lines (removed pseudocode, redundant tables, Prerequisites)
using-archeflow: 185 -> 55 lines (removed archetypes table, PDCA diagram, progress
indicators, dry-run example, full skills reference)
2026-04-06 20:43:23 +02:00
1baaa79946
refactor: simplify memory and shadow-detection skills
...
Trim verbose implementation details that duplicate what the bash helper
scripts already handle. Memory skill: 278 -> 120 lines. Shadow detection
skill: 180 -> 66 lines. All essential protocols, tables, and commands
preserved; removed redundant algorithm descriptions, multiple examples,
and narrative prose.
2026-04-06 20:43:08 +02:00
8837a359ac
refactor: simplify memory and shadow-detection skills
...
Trim verbose implementation details that duplicate what the bash helper
scripts already handle. Memory skill: 278 -> 120 lines. Shadow detection
skill: 180 -> 66 lines. All essential protocols, tables, and commands
preserved; removed redundant algorithm descriptions, multiple examples,
and narrative prose.
2026-04-06 20:42:47 +02:00
af1f4e7da7
refactor: merge attention-filters into check-phase skill
...
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.
2026-04-06 20:41:36 +02:00
e19ff0acc3
refactor: refocus ArcheFlow as workspace orchestrator, not feature-dev competitor
...
- 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
2026-04-04 18:44:18 +02:00
1bf1376a80
feat: implement archeflow-review.sh for Guardian-only diff review
...
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.
2026-04-04 18:39:06 +02:00
6309614bfa
feat: add sprint runner and review-only skills
2026-04-04 18:21:19 +02:00
6854e858a4
fix: address v0.7.0 review findings
...
- 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
2026-04-04 09:36:05 +02:00
44f0896e3c
docs: update CHANGELOG and version for v0.7.0
2026-04-04 09:36:05 +02:00
29762a8464
feat: add strategy abstraction with pdca and pipeline strategies
2026-04-04 09:36:05 +02:00
a6dcd2c956
feat: add plan granularity constraint to plan-phase and creator
2026-04-04 09:36:05 +02:00
516fe11710
feat: add evidence-gated verification to check phase and reviewers
2026-04-04 09:36:05 +02:00
f10e853d8e
feat: add structured status tokens to all agents and run skill
2026-04-04 09:36:05 +02:00
eabf13b9b0
feat: add context isolation protocol to attention-filters and all agents
2026-04-04 09:36:05 +02:00
57e95ba151
docs: add v0.4.0 changelog, update to v0.6.0
2026-04-04 08:51:19 +02:00
3c7d336c93
feat: add Explorer skip heuristic to plan-phase skill
2026-04-04 08:51:19 +02:00
12575b5a47
feat: expand attention-filters from stub to full skill
2026-04-04 08:51:19 +02:00
362fb9ada9
fix: address v0.5.0 review findings
...
- Add --to/--test-cmd mutual exclusivity guard in rollback script
- Convert all jq string interpolation to --arg (cmd_extract, cmd_inject, cmd_forget)
- Fix CRITICAL/WARNING grep to match table rows only (not prose)
- Add thorough+cycle-1 guard to fast-path bash snippet in check-phase
- Clarify prev_run_id selection comment (tail -1 = most recent non-current)
2026-04-04 08:44:16 +02:00
c3f5df8161
docs: update CHANGELOG and version for v0.5.0
2026-04-04 08:44:16 +02:00
c5174e88eb
feat: flesh out check-phase with parallel reviewer spawning protocol
2026-04-04 08:44:16 +02:00
5e2117c9be
feat: add cross-run finding regression detection
2026-04-04 08:44:16 +02:00
30ddc6a2c4
feat: add per-workflow model assignment configuration
2026-04-04 08:44:16 +02:00
008315b0c4
feat: add lib script validation at run initialization
2026-04-04 08:44:16 +02:00
d9ec148bb3
fix: address review findings (rollback mainline, audit flag, confidence gate, test pattern, jq args)
...
- git revert uses --mainline 1 for merge commits
- inject dispatch passes all args so --audit flag is reachable
- confidence gate defaults to 0.0 (triggers gate) instead of 0.7 (bypasses)
- test-first grep uses word-boundary patterns to avoid false positives
- jq uses --arg instead of string interpolation for safe filtering
2026-04-04 08:30:43 +02:00
dd82944529
feat: add memory injection audit trail for effectiveness tracking
2026-04-04 08:30:31 +02:00
8af9db2c12
feat: add test-first validation gate in Do phase
2026-04-04 08:30:31 +02:00
7f99d52a09
feat: add archeflow-rollback.sh for post-merge test failure auto-revert
2026-04-04 08:30:31 +02:00
34f101c166
feat: specify confidence gate parsing and mini-Explorer spawning in run skill
2026-04-04 08:30:31 +02:00
960aba5faa
feat: clarify worktree merge flow with explicit git commands in run skill
2026-04-04 08:30:31 +02:00
2247e52ae4
fix: unify feedback routing tables across orchestration, act-phase, artifact-routing
2026-04-04 08:30:31 +02:00
6bc5e48357
feat: flesh out do-phase skill from stub to full execution protocol
2026-04-03 12:25:29 +02:00
6b0a9b7b90
feat: add presence skill — visible value, not noise
...
Defines how ArcheFlow communicates: compact status lines per phase,
show outcomes not mechanics, silence for clean passes. One-line
activation indicator at session start.
2026-04-03 12:18:18 +02:00
9e22ff5822
docs: rewrite README, add CHANGELOG, update skill index and roadmap
...
- README: complete rewrite with all 24 skills, 8 scripts, architecture overview
- CHANGELOG: v0.1.0 → v0.2.0 → v0.3.0 with full feature history
- using-archeflow: updated skill reference to all 24 skills in 6 categories
- plugin.json: version bump to 0.3.0
- roadmap: updated with v0.3.0 features
2026-04-03 12:05:53 +02:00
ee5dfa70b8
feat: add multi-project orchestration with dependency DAG and shared budget
2026-04-03 11:41:06 +02:00
19f8f76232
feat: add memory, convergence, colette bridge, templates, progress, effectiveness, git integration
...
- skills/memory: cross-run learning from recurring findings + lib/archeflow-memory.sh
- skills/convergence: oscillation detection + early termination in multi-cycle runs
- skills/colette-bridge: auto-inject voice profiles, personas, characters from colette.yaml
- skills/templates: workflow/team/archetype gallery with init/save/share
- skills/progress: live .archeflow/progress.md during runs
- skills/effectiveness: per-archetype signal-to-noise + cost efficiency scoring
- skills/git-integration: auto-branch per run, commit per phase, rollback support
2026-04-03 11:40:04 +02:00
b6df3d19fd
feat: add automated PDCA loop, domain adapters, cost tracking, DAG renderer
...
- skills/run: automated PDCA execution loop with --start-from, --dry-run
- skills/artifact-routing: inter-phase artifact protocol with context injection
- skills/act-phase: structured review→fix pipeline with cycle feedback
- skills/domains: domain adapter system (writing, code, research)
- skills/cost-tracking: per-agent cost estimation, budget enforcement
- lib/archeflow-dag.sh: ASCII DAG renderer from JSONL events
- lib/archeflow-report.sh: updated with DAG section, cycle diff, --dag/--summary flags
2026-04-03 11:20:14 +02:00
1753e69a9f
feat: add process logging with DAG-based event sourcing
...
Event-sourced orchestration logging: JSONL events with parent
relationships form a DAG for causal reconstruction of agent flows.
Includes bash event emitter (jq-based) and markdown report generator.
2026-04-03 11:06:02 +02:00
8dec44d199
chore: consolidate plugin for sharing
...
- Add .gitignore (ignore .archeflow/ runtime state)
- Move custom archetype examples from .archeflow/ to examples/
- Remove onboarding skill (covered by using-archeflow + README)
- Remove internal planning doc
- Clean roadmap (remove tool.archeflow references)
- Simplify session-start hook config
- Polish README for external users
2026-04-03 07:29:53 +02:00
5eefa309cb
fix: redesign adaptation rules per Guardian review — resolve race conditions
...
Guardian review (Sprint 1) found critical issues in adaptation rules:
- A2 couldn't skip already-spawned parallel reviewers (race condition)
- A1 escalated then A2 skipped the escalated reviewers (self-defeating)
- A3 "Pause" conflicted with parallel agent execution
- A1's "thorough" override was unreachable dead code
Fixes:
- Reorder rules: A3 (after Plan) → A2 (after Guardian) → A1 (after Check)
- Guardian now runs first in Check phase, A2 evaluated before spawning others
- A2 doesn't apply to escalated workflows or first cycle of thorough
- A3 task-understanding pause runs before Do phase (no cancellation needed)
- Added explicit evaluation order diagram
2026-04-03 06:25:27 +02:00
8755d68dc9
fix: address Guardian review findings — sync Creator agent, wire hooks, add cost table
...
- Sync agents/creator.md output format with plan-phase skill (3-axis confidence,
alternatives considered, mini-reflect) — fixes adaptation rule A3 dependency
- Wire hook points into orchestration Act phase (pre-merge, post-merge)
- Add cost-per-agent-tier table to autonomous-mode for budget scheduling
- Add team preset loading reference to orchestration Step 0
2026-04-03 06:23:47 +02:00
761d64b821
feat: DX, extensibility, and adaptive intelligence — completes 25-feature backlog
...
Sprint 5 — Developer Experience (D1-D5):
- Progress indicators with emoji phase markers during orchestration
- Dry-run mode: preview workflow without executing
- /archeflow:status and /archeflow:history commands
- Onboarding skill for first-time users
Sprint 6 — Extensibility (E1-E4, A4-A6):
- Archetype composition: combine 2 archetypes into super-reviewer
- Team presets: save common team configs in .archeflow/teams/
- Hook points: pre-plan, post-check, pre-merge, post-merge custom validation
- Workflow template library: api-design, migration, dep-upgrade, docs, hotfix
- Reviewer profiles: per-project config of which reviewers run
- Explorer cache: skip redundant research on recently-explored code
- Learning from history: track archetype usefulness, recommend profile changes
2026-04-03 06:20:44 +02:00
83e09b70f2
feat: workflow intelligence, quality loop, completion promises, parallel teams
...
Sprint 1 — Workflow Intelligence (A1-A3):
- Conditional escalation: fast→standard on 2+ CRITICALs
- Guardian fast-path: skip remaining reviewers on clean pass
- Confidence-triggered escalation: pause/upgrade/probe on low scores
Sprint 2 — Quality Loop (B1-B2, B5-B6):
- Maker self-review checklist before submitting to Check phase
- Proposal diff ("What Changed") on cycle 2+ revisions
- Convergence detection: escalate to user if same finding persists 2 cycles
- Cross-archetype dedup: merge duplicate findings from different reviewers
Sprint 3 — Completion & Verification (B3-B4):
- Completion promise: user-defined done criteria checked in Act phase
- Post-merge verification: run tests on main, auto-revert on failure
Sprint 4 — Parallel & Scale (C1-C4):
- Parallel team orchestration: 2-3 independent teams with merge gate
- Task dependency graph in autonomous queue format
- Auto-resume on interruption via .archeflow/state.json
- Budget-aware scheduling with automatic workflow downgrade
2026-04-03 06:17:53 +02:00
5139f1ad89
feat: Ralph Loop integration — mini-reflect, alternatives, structured confidence
...
- Mini-Reflect for fast workflow: Creator must restate task, list assumptions,
name highest-damage risk before proposing (catches misunderstandings early)
- Alternatives Considered section: Creator must evaluate 2+ approaches with
rejection rationale before committing to one (prevents tunnel vision)
- Structured confidence scoring: 3-axis table (task understanding, solution
completeness, risk coverage) replaces bare 0.0-1.0 number. Low scores
trigger targeted action (clarify, upgrade workflow, or research)
- Mini-Reflect fallback for skipped tasks: quick reflection even when
ArcheFlow doesn't activate (non-trivial single-file changes)
2026-04-03 06:08:01 +02:00
d08dc657d1
feat: core improvements — feedback loop, attention filters, shadow heuristics, metrics, auto-activation
...
- Cross-cycle feedback protocol with structured finding format, routing, and resolution tracking
- Attention filter enforcement: explicit context include/exclude per archetype
- Shadow detection: quantitative checklists with concrete thresholds
- Orchestration metrics: per-phase timing, agent count, findings summary
- Autonomous mode wiring: checkpoint protocol, session log, stop conditions
- Auto-activation: SessionStart hook fires ArcheFlow for implementation tasks without user config
- Emoji avatars for all 7 archetypes
- Standardized finding format across all reviewers for cross-cycle tracking
- Persisted implementation plan in docs/
2026-04-03 06:02:10 +02:00
ebe943a67e
refactor: realistic shadows for Claude behavior
...
- Creator: Perfectionist → Over-Architect (Claude over-designs, doesn't endlessly revise)
- Maker: Cowboy → Rogue (same behavior, better name)
- Trickster: Saboteur → False Alarm (Claude floods with noise, doesn't sabotage)
2026-04-02 18:50:40 +00:00
