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 skills add ychampion/cskill-agents --skill concurrency-safe-tool-batchinggit clone --depth 1 https://github.com/ychampion/cskill-agentsWrote 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.
[](https://agentmods.dev/skills/ychampion/cskill-agents/concurrency-safe-tool-batching)<a href="https://agentmods.dev/skills/ychampion/cskill-agents/concurrency-safe-tool-batching"><img src="https://agentmods.dev/badge/skills/ychampion/cskill-agents/concurrency-safe-tool-batching/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.
<a href="https://agentmods.dev/skills/ychampion/cskill-agents/concurrency-safe-tool-batching"><img src="https://agentmods.dev/badge/skills/ychampion/cskill-agents/concurrency-safe-tool-batching.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00029 | $0.00439 |
| Opus 5 | $0.00015 | $0.00219 |
| Sonnet 5 | $0.00006 | $0.00088 |
| Haiku 4.5 | $0.00003 | $0.00044 |
Grade A, and why
concurrency-safe-tool-batching 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 10d 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.
How it starts
The opening of the file, as written. The whole thing — 28 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SKILL: Concurrency-Safe Tool Batching
Domain: tool-orchestration
Trigger: Use when one turn contains multiple tool calls and some are safe to run together while others mutate state or require exclusive execution.
Source Pattern: Distilled from reviewed tool execution, streaming, persistence, and output-budget implementations.
Core Method
Partition the requested tool calls into successive batches where each batch is either a set of read-only calls that may run together or a single exclusive call that must run alone. Decide concurrency safety from each tool’s declared behavior and validated inputs, not from naming conventions. Execute safe batches in parallel, then apply any shared-context updates only after the whole batch completes. Run mutating or stateful tools one at a time so ordering and side effects remain deterministic.
Key Rules
- Treat a tool as concurrency-safe only after validating its input and checking its declared safety rules; if anything is ambiguous, fall back to serial execution.
- Append queued context modifiers from concurrent batches after they finish so the shared tool use context stays consistent before the next batch.
- Always reorder only read-only batches; non-read-only batches must run on their own and update context inline so ordering and side effects remain deterministic.
Example Application
When building a CLI tool runner, first group together safe read-only lookups like file reads or searches, run that group in parallel, then execute any edit, write, or stateful shell step on its own. This gives you parallelism where it is safe without letting mutating tools race each other.
Anti-Patterns (What NOT to do)
- Assume all tools are concurrency-safe and run them in parallel, which can break stateful tools like shell edits.
- Drop queued context modifiers from concurrent batches and let subsequent batches run on stale context, leading to permission and session problems.
- Merge tool use results out of order when non-read-only tools follow concurrent batches, which breaks user expectations about sequential effects.
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.
- 10d ago First seen · 28 lines · 29 tokens per session scan A 0633c7778427
concurrency-safe-tool-batching is a skill published in the GitHub repository ychampion/cskill-agents (36 stars, last pushed 5mo ago), licensed MIT. It adds 29 tokens to every session and 439 once invoked, about $0.0001 per session on Opus 5. 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 2026-08-30.
Other skills, from other repositories
grape
Use Grape MCP for Codex context continuity in coding repositories. Use when a task needs repeated-turn context, omitted context restore, stale-context checks, invalidation checks, or safe continuity across branch and dirty-worktree changes.
review
Validate plans, execution, or PRs against wish criteria — returns SHIP / FIX-FIRST / BLOCKED with severity-tagged gaps.
work
Execute an approved wish plan — orchestrate subagents per task group with fix loops, validation, and review handoff.
brainstorm
Explore ambiguous or early-stage ideas interactively — tracks wish-readiness and crystallizes into a design for wish.
genie
Entry point for Genie operations — routes bug reports, questions, and operational commands, resumes existing lifecycle state, and orchestrates work that needs durable planning or coordination. Other ordinary requests bypass the lifecycle with a one-line notice unless the user asks for Genie.
wish
Convert an idea into a structured wish plan with scope, acceptance criteria, and execution groups for work.