Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/moneytosms/claude-start/batchgit clone --depth 1 https://github.com/moneytosms/claude-startWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.00270 |
| Opus 5 | $0.00000 | $0.00135 |
| Sonnet 5 | $0.00000 | $0.00054 |
| Haiku 4.5 | $0.00000 | $0.00027 |
Grade A, and why
batch scanned grade A with 0 findings 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 2d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
/batch
Split a large task into parallelisable subtasks and delegate to subagents.
Steps
- Decompose the task into independent subtasks (no shared write targets between subtasks)
- For each subtask, identify:
- What it does
- Which files it reads/writes (must be disjoint from other subtasks)
- Which agent is appropriate (ReadOnly, BuildValidator, etc.)
- What the expected output is
- Ask permission before spawning any agent — list the subtasks and agents first
- On approval, spawn one subagent per subtask with a self-contained prompt
- Collect all outputs
- Merge results: resolve any conflicts, integrate findings, report summary
Rules
- Never spawn agents without explicit permission
- Subtask write scopes must be disjoint — overlapping writes will corrupt state
- Include all necessary context in each subagent prompt (they have no shared memory)
- If a subtask fails, report it without blocking others that succeeded
- Prefer 2-4 subtasks. If you need more than 6, the task decomposition is too fine-grained.
Use cases
- Large refactors touching many files
- Multi-file migrations
- Parallel research + implementation
- Running tests while generating documentation
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.
- 2d ago First seen · 32 lines · 0 tokens per session scan A 1d3f373c3e54
batch is a command published in the GitHub repository moneytosms/claude-start (2 stars, last pushed 10d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 270 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
feature-orchestrate
Autonomous multi-phase feature execution.
track
Show progress, flags and open decisions.
plan-status
Show feature plan status for the current workspace.
plan
Plan a feature and write its artifact tree.
feature-resume
Resume implementation on an existing feature, interactively.
review
Check implementation against the feature ledger.