From majestic-engineer
Tracks task execution attempts in epics with retry context, failure gating to prevent loops, progress persistence, and structured receipts for resumption.
npx claudepluginhub majesticlabs-dev/majestic-marketplace --plugin majestic-engineerThis skill uses the workspace's default tool permissions.
The attempt ledger tracks execution attempts for tasks, enabling:
Autonomously executes epics by orchestrating waves of swarm workers to implement issues, close tasks, and generate code changes until completion. Ideal for large-scale autonomous coding.
Executes pending Claude Code Tasks in dependency order with wave-based concurrent execution, adaptive verification, retries, task group filtering, and configurable parallelism.
Orchestrates multi-day execution of complex tasks via milestone pipelines with plan-crafting, run-plan, review-work phases, checkpoints, and recovery.
Share bugs, ideas, or general feedback.
The attempt ledger tracks execution attempts for tasks, enabling:
# .agents/relay/epic.yml
version: 2
id: "20260111-feature-name"
source: "docs/plans/xxx.md"
created_at: "2026-01-11T17:30:00Z"
title: "Feature Name"
description: |
Brief description of the epic.
tasks:
T1:
title: "Create users migration"
priority: p1
points: 2
files:
- db/migrate/xxx_create_users.rb
depends_on: []
acceptance_criteria: # Generated by relay:init if missing
- "Migration creates users table with email and password_digest"
- "Migration is reversible"
- "Email column has unique index"
# .agents/relay/attempt-ledger.yml
version: 1
epic_id: "20260111-feature-name"
started_at: "2026-01-11T17:30:00Z"
settings:
max_attempts_per_task: 3
timeout_minutes: 15
task_status:
T1: completed
T2: in_progress
T3: pending
T4: blocked
attempts:
T1:
- id: 1
started_at: "..."
ended_at: "..."
result: success
receipt:
summary: "Created migration and model"
files_changed: [file1.rb, file2.rb]
T2:
- id: 1
result: failure
receipt:
error_category: missing_dependency
error_summary: "Database not migrated"
suggestion: "Run db:migrate first"
gated_tasks:
T5:
reason: "max_attempts_exceeded"
gated_at: "..."
Tasks are gated (blocked from retry) when:
| Condition | Gating Action |
|---|---|
attempts >= max_attempts | Gate with max_attempts_exceeded |
| Same error repeated 2+ times | Gate with repeated_failure |
| Quality gate returns BLOCKED | Gate with quality_gate_blocked |
| Manual block by user | Gate with user_blocked |
receipt:
summary: "Brief description of what was done"
files_changed: [list, of, files]
quality_gate_verdict: APPROVED
learning: "Reusable insight from this task" # Compound learning
pattern_tags: [yq, shell, quoting] # For aggregation
receipt:
error_category: missing_dependency | code_error | test_failure | quality_gate | cli_error
error_summary: "What went wrong"
quality_gate_verdict: NEEDS CHANGES | BLOCKED # If quality gate failed
quality_gate_findings: | # Structured findings for retry context
## Finding 1: Missing test coverage
**Severity:** HIGH
**File:** app/models/user.rb:15
**Fix:** Add unit tests for validation
suggestion: "What to try differently"
learning: "What was learned from the failure" # Compound learning
pattern_tags: [migration, database] # For aggregation
Learnings are extracted after each task attempt, aggregated at epic completion, then cleared.
After each attempt (success or failure), a lightweight prompt extracts:
When epic completes, the learning-processor agent:
pending → in_progress → completed
↓
failure → pending (retry)
↓
max_attempts → gated
Each fresh Claude instance receives:
This prevents context drift and ensures each attempt builds on prior learnings.
| Command | Purpose |
|---|---|
/majestic-relay:init | Create epic and ledger from blueprint |
/majestic-relay:status | Show progress and gated tasks |
/majestic-relay:work | Execute tasks with attempt tracking |
Settings are stored in attempt-ledger.yml at init time:
max_attempts_per_task: 3 - Retry limit before gatingtimeout_minutes: 15 - Max execution time per taskOverride at runtime with --max-attempts N flag.