MVPavan/via / .claude/skills/beads
beads
VerifiedUse for Beads dependencies, intake states, recovery, workstream tracking, or unfamiliar task-lifecycle operations. Routine tracking uses AGENTS.md and runtime context.
Beads
Use Beads as the durable work tracker, following .beads/beads.md. Routine
tracking follows AGENTS.md and supplied runtime context without loading this
reference workflow. Recover missing context with bd prime; if empty, use
bd where. For command syntax, consult references/commands.md or current help.
Repository conventions
- Attribute every write with
--actor "<runtime>:<session-or-purpose>". Labels use lowercase kebab-case; priority is 0–4, with 0 critical. - Reuse the work item's record. Create independent children only when the work benefits from separate ownership or acceptance; add actual prerequisite edges.
- Check for a governing spec before creating an epic.
--spec-idcarries the spec,--designthe roadmap, and notes carryplan: <path>. Explain absence of a spec in the issue. Spec IDs do not inherit to children. - Phase epic titles start
[<phase-id>]and carryws-<name>; stage acceptance must match the roadmap's Verify contract. Intake, idea and backlog states are defined in.beads/beads.md, not invented by each workflow. - Close only after acceptance holds, with verification evidence in
--reason. Generated workstream mirrors come fromscripts/bd-render-tracking.sh; missing tooling is a reported limitation, not permission to hand-edit mirrors.
Explicit repository overrides
Beads stores work, not general persistent knowledge. This repository deliberately
chose file-based memory over bd remember; preserve that choice unless changed.
Write persistent memory only when explicitly requested, and follow any higher-
priority runtime instructions governing its destination.
Use conservative Git authority. Session close normally refreshes the Beads export when records changed; an explicit user restriction on writable paths takes precedence, so report a deferred export rather than editing outside that scope.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 467k | Updated | Claude CodeCodex | Verified | Install |
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 467k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 319k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 319k | Updated | Claude CodeCodex | Verified | Install |