AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified Apache-2.0 Self-run

Next

skill-yassimba-loom-next · by Yassimba

Claim the next actionable Beads task using bv and br. Use when the user asks what to work on next or asks to pick, grab, claim, or start the next task.

No reviews yet
0 installs
0 views
view→install

Install

$ agentstack add skill-yassimba-loom-next

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No 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 Used
  • 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.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-yassimba-loom-next)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
yesterday

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

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 →
Are you the author of Next? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Next

Select one ready task from the Beads dependency graph, claim it for this session, and start it. Run every br command through this skill's actor-aware launcher:

  • macOS/Linux: /scripts/br-agent.sh
  • Windows: & "\scripts\br-agent.ps1"

Replace ` with the directory containing this SKILL.md. The launcher resolves one session-unique actor, passes it to br --actor, preserves all arguments and the exit code, and fails before br` runs when it cannot identify the session.

  1. Confirm the repository contains .beads/ and both br and bv are on PATH. If any prerequisite is missing, report it and stop.
  2. Run bv --robot-next. Use only bv --robot-* commands; bare bv opens an interactive TUI.
  3. Accept only the returned top pick with a non-empty claim_command. Verify its current state with show --json: it must be open, ready, and unassigned. An assignment to another actor is a live claim.
  4. Claim it atomically with update --claim. If the claim loses a race, rerun the selection from step 2; never force-reclaim another actor's task.
  5. Run show --json again and read the full task. State the claimed ID and title, then treat its body as the current request and begin the work using the repository's normal workflow and relevant skills.

If bv returns no task, run ready --json. Report an empty list as an empty frontier; list epics as a ready epic frontier; report any non-epic as a bv discrepancy. Use the fallback only for diagnosis. Claim only a bv top pick.

Actor adapters

BEADS_ACTOR is the launcher's explicit, stable input and always wins. CLAUDE_CODE_SESSION_ID, CODEX_SESSION_ID, CODEX_THREAD_ID, OPENCODE_SESSION_ID, and PI_SESSION_ID are host adapters in that order. Loom's OpenCode plugin supplies OPENCODE_SESSION_ID from the documented shell.env session input; the other hosts do not provide a shared stable contract. OpenAI's stable Codex environment-variable list does not include either Codex session identifier. Add new host adapters only when they identify one session uniquely. A human login name and a per-command process ID are not session identities.

Detected actors keep the complete raw session ID after a provider prefix. Strip only that prefix to resume the owning session:

| Beads assignee | Resume | | --- | --- | | claude- | claude --resume | | codex- | codex resume | | opencode- | opencode --session | | pi- | pi --session |

Concurrent instances of one agent stay separate because each launcher process reads its own session ID. If the host does not expose a unique ID, stop and ask the user to set BEADS_ACTOR=- explicitly.

Pass an explicit override through BEADS_ACTOR; the launcher rejects --actor arguments so the real br process receives exactly one actor.

Done when exactly one ready task is claimed for the resolved actor and its work has begun, or when a named stop condition has been reported.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.