execution-tracking

Bootstrap and operate the bd CLI as an external task tracker for agent execution plans.

Updated Dec 20, 2025
One-click install
npx skills add https://github.com/cristoslc/LLM-personal-agent-patterns --skill execution-tracking-cristoslc
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: execution-tracking
Source: https://github.com/cristoslc/LLM-personal-agent-patterns/tree/main/L3-agents-core/.agents/skills/execution-tracking
Command: npx skills add https://github.com/cristoslc/LLM-personal-agent-patterns --skill execution-tracking-cristoslc

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? Built-in agent todo lists vanish between sessions and cannot be shared across agent runtimes or observed by supervisors. This Skill installs and operates the bd (beads) CLI as a persistent, git-backed source of truth for implementation plans, tasks, dependencies, and progress. ## Core Features & Use Cases - Plan and task registration: Translates spec-management intent (implementation plans, tasks, origin refs, spec tags) into concrete bd commands, creating epics with --external-ref and child tasks with spec: labels. - Dependency-aware work selection: Uses bd ready and bd blocked to surface truly claimable work, claim tasks atomically, and close or abandon them with queryable reasons. - Escalation and fallback: Defines an abandon-and-escalate protocol back to spec-management, plus a JSONL/Markdown ledger fallback when bd cannot be installed. - Use Case: A SPEC comes up for implementation — the Skill bootstraps bd, creates an epic linked via origin ref, ingests a superpowers plan file with scripts/ingest-plan.py, wires sequential task dependencies, and keeps status visible to external observers. ## Quick Start Ask the agent to create a tracked implementation plan for SPEC-003 and show what work is ready to claim next.

Frequently Asked Questions about execution-tracking

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a tracked implementation plan with bd?▼

Create an epic with `bd create "Title" -t epic --external-ref SPEC-003 --json`, then add child tasks with `--parent <epic-id>` and `--labels spec:SPEC-003`. Always use `--json` to capture the assigned issue IDs for later commands.

How do I find the next task to work on with bd?▼

Run `bd ready --json`, which evaluates actual dependency chains to list unblocked, claimable tasks. Do not use `bd list --ready` — it only filters by status and ignores blockers.

What happens if the bd CLI cannot be installed?▼

The Skill falls back to a neutral text task ledger in JSONL or Markdown format, controlled by the `fallback_format` setting in `.agents/execution-tracking.vars.json`. It keeps the same status model: open, in_progress, blocked, closed.

Does bd require a Dolt server to run?▼

No. bd uses SQLite by default and needs no server. Dolt is optional for remote sync and is only used when `use_dolt` is set to true in the Skill's configuration file.

Why should I avoid bd edit when updating tasks?▼

`bd edit` opens an interactive editor like vim or nano, which blocks non-interactive agents. Use `bd update <id>` with inline flags such as `--status`, `--notes`, or `--add-label` instead.

When should execution tracking not be used directly on an artifact?▼

Coordination-tier artifacts (EPIC, VISION, JOURNEY) should not be tracked directly. Spec-management must first decompose them into implementable children (SPECs or STORYs), which this Skill then tracks.