Install
$ agentstack add skill-trebormc-drupal-ai-agents-beads-task-tracking ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Environment
Beads runs in its own DDEV container (ddev-{project}-beads). A wrapper script at /usr/local/bin/bd is installed in your container, so all bd commands work directly -- they are transparently delegated to the Beads container via SSH.
The .beads/ directory lives in the project root (/var/www/html/.beads/) and is shared across all containers through the project volume.
Session Start (ALWAYS DO THIS FIRST)
# 1. Initialize Beads if not already done
ls -la .beads/ 2>/dev/null || bd init --quiet
# 2. Get context from previous sessions (prints a project/task summary — if it
# prints without error, Beads is working)
bd prime
# 3. See what tasks are ready
bd ready --json
If bd fails
| Symptom | Cause | Fix | |---------|-------|-----| | bd: command not found or SSH errors | Beads container down | Ask the user to run ddev restart on the host | | Unknown subcommand / flag error | Wrong syntax | Check available commands with bd --help | | Command hangs | You ran bd edit | Never use bd edit — use bd update --flags |
Creating Tasks
bd create is NOT idempotent — running it twice creates two tasks. Before creating, check the task does not already exist (bd ready --json or bd list --json). If a previous create attempt failed ambiguously (e.g., SSH error), verify with bd ready --json before retrying.
# Create a task with priority (P0-P3)
bd create "Implement user authentication" -p 1 --json
# Create a subtask under a parent
bd create "Add validation for edge case" -p 2 --parent --json
# Create multiple tasks for a feature
bd create "Create service class" -p 1 --json
bd create "Add form handler" -p 1 --json
bd create "Write unit tests" -p 2 --json
Priority Levels
| Level | Meaning | When to use | |-------|---------|-------------| | P0 | Critical | Blockers, security issues, broken builds | | P1 | High | Core features, main deliverables | | P2 | Medium | Normal tasks, enhancements | | P3 | Low | Nice-to-haves, polish, cleanup |
Working on Tasks
# Mark a task as in progress
bd update --status in_progress
# Add progress notes as you work
bd update --notes "Implemented service, working on form validation"
# Close when done
bd close --reason "Implemented and all tests passing" --json
Querying Tasks
# List all ready tasks
bd ready --json
# Show details of a specific task
bd show --json
Session End ("Land the Plane") - MANDATORY
Before finishing any work session, ALWAYS complete these steps:
# 1. File follow-up tasks for remaining work
bd create "TODO: Add integration tests" -p 2 --json
# 2. Close all completed tasks
bd close --reason "Done: implemented service and unit tests" --json
# 3. Document paused state for in-progress tasks
bd update --notes "Paused at: form validation logic, needs edge case handling"
Quick Reference
| Action | Command | |--------|---------| | Initialize | bd init --quiet | | Get context | bd prime | | List ready | bd ready --json | | Create task | bd create "Title" -p 1 --json | | Create subtask | bd create "Title" -p 2 --parent --json | | Show task | bd show --json | | Mark active | bd update --status in_progress | | Add notes | bd update --notes "..." | | Close task | bd close --reason "Done" --json |
Integration with Ralph Loop
When running inside Ralph Loop ([RALPH LOOP - Iteration X]):
Planning Phase (Iteration 1):
# Read requirements and create tasks
bd create "Task from requirements" -p 1 --json
bd create "Another task" -p 2 --json
# Signal completion
echo "PLANNING_COMPLETE"
Execution Phase (Iterations 2+):
# Get next task
bd ready --json
# Work on it
bd update --status in_progress
# ... do the work ...
bd close --reason "Completed" --json
# Create new subtasks if discovered
bd create "New discovered task" -p 2 --json
The loop exits automatically when bd ready returns an empty array [].
Rules
- NEVER use
bd edit-- it opens an interactive editor that will hang - NEVER run git commit or git push -- leave all git operations to the user
- Always use
--jsonflag on create, close, and show for parseable output - Always close tasks with a reason -- this documents what was accomplished
- Reference task IDs in commit messages -- suggest format:
"Add feature X (bd-a1b2)"
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: trebormc
- Source: trebormc/drupal-ai-agents
- License: Apache-2.0
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.