bead-recovery-specialist

bead-recovery-specialist is an agent for Claude Code from jeremylongshore/dolt-mcp-vcs-plugin. It costs 88 tokens per session (1,225 once invoked), scanned A, original, Apache-2.0.

A recovery agent for Beads (bd) and Dolt, a database system used by bd. It diagnoses server-start failures, leftover servers, possible lost writes, stale JSONL exports, and moves between embedded and server modes.

In plain words
What is it for?
Use it to inspect bd and Dolt health, flush data with `bd export`, identify server problems, and prepare safe commands for recovery or migration.
Why use it?
It helps investigate database incidents while reducing the risk of losing work. Heavy recovery actions are presented for a human to approve and run, rather than being executed automatically.

Agent for Claude Code

Written for Claude Code: background in frontmatter. Also seen: model in frontmatter.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the dolt-mcp-vcs plugin — 1 skill, 5 agents, 1 MCP server shipped together

Good fit Use it to inspect bd and Dolt health, flush data with bd export, identify server problems, and prepare safe commands for recovery or migration.

Compare 6 agents from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add jeremylongshore/dolt-mcp-vcs-plugin
Claude Code
/plugin install dolt-mcp-vcs

Made for: Claude Code.

Or install dolt-mcp-vcs, the plugin that ships this one along with the rest of its 1 skill, 5 agents, 1 MCP server.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for bead-recovery-specialist

README.md
[![agentmods](https://agentmods.dev/badge/agents/jeremylongshore/dolt-mcp-vcs-plugin/bead-recovery-specialist/github.svg)](https://agentmods.dev/agents/jeremylongshore/dolt-mcp-vcs-plugin/bead-recovery-specialist)
Your own site
<a href="https://agentmods.dev/agents/jeremylongshore/dolt-mcp-vcs-plugin/bead-recovery-specialist"><img src="https://agentmods.dev/badge/agents/jeremylongshore/dolt-mcp-vcs-plugin/bead-recovery-specialist/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for bead-recovery-specialist

Your own site · 80×15
<a href="https://agentmods.dev/agents/jeremylongshore/dolt-mcp-vcs-plugin/bead-recovery-specialist"><img src="https://agentmods.dev/badge/agents/jeremylongshore/dolt-mcp-vcs-plugin/bead-recovery-specialist.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 88 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,225 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00088 $0.01225
Opus 5 $0.00044 $0.00613
Sonnet 5 $0.00018 $0.00245
Haiku 4.5 $0.00009 $0.00122

Measured 12d ago against content hash 62f71d8cd9fa, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

bead-recovery-specialist scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

tools: Read, Bash(bd export:*), Bash(bd version:*), Bash(bd dolt show:*), Bash(bd dolt status:*), Bash(bd config get:*), Bash(bd config list:*), Bash(bd --help:*), Bash(bd dolt --help:*), Bash(dolt status:*), Bash(curl:*
agents/bead-recovery-specialist.md · 51 lines

How it starts

The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a bd and Dolt recovery specialist. You stabilize a broken or sprawled bd Dolt backend without losing data.

Mutation safety — recommend, don't execute (blueprint §3). Your safe direct actions are read-only diagnostics (bd dolt show/status, server-health.sh), the JSONL flush (bd export), and idle-server reaping (non-destructive — bd respawns). The heavier recovery levers — bd dolt killall, bd backup sync, bd config set, dolt reset, embedded↔server migration — are recommend-only: surface the exact command for the operator (they are denied to you), explain the rollback, and let the human run it. Never run a destructive recovery step before a verified bd export flush.

Fetch the current truth — don't recall it. You run in your own context, so before asserting any version-specific behavior, read it live: run bd --help / bd <cmd> --help / bd dolt show, check the installed version (bd version), or curl the upstream CHANGELOG / issue. references/dolt-internals.md is only a directory of authoritative sources. Re the "rapid-write race" (upstream failure mode 6): verify its status against the installed binary's behavior + the upstream CHANGELOG/issue before pronouncing — as of recent bd it is reported fixed at the SQL-transaction level (DB writes atomic + retried), with residual .beads/issues.jsonl lag from the export throttle. Confirm, then advise; the installed binary is the authority.

Core Responsibilities

  1. Triage dolt-server incidents (won't start, orphaned, port churn).
  2. Resolve JSONL-lag confusion — distinguish the throttle from data loss.
  3. Migrate workspaces between embedded and server mode safely; reap idle servers to tame sprawl.
  4. Clean up orphaned servers and port sprawl.

Process

  1. Checkpoint first. Before any change, bd export then bd backup sync — never operate without a rollback point.
  2. Inventory. Run bash ${CLAUDE_PLUGIN_ROOT:-.}/scripts/server-health.sh to map running servers to workspaces and detect sprawl.
  3. JSONL lag. If JSONL looks stale after a burst, it is the 60s export throttle, not loss. Flush with bd export; for gitignored .beads, set bd config set export.interval 1s. Confirm the DB is correct via bd dolt show and a row count.
  4. Server won't start / orphaned. Check bd dolt status; inspect dolt-server.pid/.port/.lock; use bd dolt killall (repo-scoped, refuses external/other-repo servers) then let bd auto-restart.
  5. Tame sprawl. Reap idle servers with bash ${CLAUDE_PLUGIN_ROOT:-.}/scripts/dolt-idle-reaper.sh --dry-run then without --dry-run — bd respawns each on its next command, so nothing is lost (the lightweight option). For a durable single-server setup, shared-server consolidation also exists; read bd init --help / bd dolt --help live for the current flags before recommending it.

Read the full file on GitHub · 51 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 12d ago First seen · 51 lines · 88 tokens per session scan A 62f71d8cd9fa

Subscribe to this mod's changes

bead-recovery-specialist is an agent published in the GitHub repository jeremylongshore/dolt-mcp-vcs-plugin (4 stars, last pushed 17d ago), licensed Apache-2.0. It adds 88 tokens to every session and 1,225 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other agents, from other repositories

ash-query-optimizer

Ash query optimizer — detects N+1 loads, suggests aggregates over load+Enum, identifies calculation vs load tradeoffs. Use when reviewing Ash queries, LiveView data loading, or domain action efficiency.

oliver-kriska/claude-elixir-phoenix · 45 tokens

d1-debugger

Autonomous diagnostic agent that investigates Cloudflare D1 database issues through 9-phase analysis (config, migrations, queries, bindings, errors, limits, performance, Time Travel, report). Use when encountering D1 query errors, migration failures, binding issues, performance degradation, or limit/quota errors.

secondsky/claude-skills · 65 tokens

performance-optimizer

Performance optimization expert. Identifies N+1 queries, memory leaks, and slow queries.

sd0xdev/sd0x-harness · 22 tokens

doctrine-performance-optimizer

Read-only performance audit of Doctrine usage: N+1 queries, fetch modes, batch processing, missing indexes, and caching opportunities. Use proactively after adding entities, relations, repository queries, or when a page/endpoint is reported slow.

dev-toolings/superpowers-symfony · 53 tokens

db-performance-tuner

Database bottleneck specialist for the Hydraia pipeline. Evidence-first tuning across PostgreSQL, MySQL/MariaDB, MongoDB, and Redis — EXPLAIN plans, index design, N+1 detection, lock contention, pooling. Dispatched by the perf/DB route at design time. Read-only; proposes changes as plan tasks. NOT a Phase 5 reviewer…

jdanigo/hydraia · 86 tokens

n1-hunter

Use this agent when you need to statically find N+1 query patterns in ORM or data-access code, loops that lazily load a relation per row, missing eager-loading (selectrelated/prefetchrelated/joinedload/includes/with/Include/Preload), or serializer-triggered per-row queries. Trigger phrases include "N+1", "N plus one"…

sigistry/marketplace · 0 tokens