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 LandonSchropp/agent-toolkit --skill dispatchgit clone --depth 1 https://github.com/LandonSchropp/agent-toolkitWrote 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/landonschropp/agent-toolkit/dispatch)<a href="https://agentmods.dev/skills/landonschropp/agent-toolkit/dispatch"><img src="https://agentmods.dev/badge/skills/landonschropp/agent-toolkit/dispatch/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/landonschropp/agent-toolkit/dispatch"><img src="https://agentmods.dev/badge/skills/landonschropp/agent-toolkit/dispatch.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 | $0.00039 | $0.01213 |
| Opus 5.5 | $0.00016 | $0.00485 |
| Sonnet 5.5 | $0.00008 | $0.00243 |
| Haiku 4.5 | $0.00004 | $0.00121 |
Grade A, and why
dispatch 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 today.
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 — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dispatch
This session delegates. It does not implement.
Concretely: create no files, change no files, in any repository, including through Bash. Edit and Write are withheld on the turn this skill loads, but they come back on the next one and Bash was never withheld at all, so the rule has to hold on its own. A task that looks small is exactly the one you will be tempted to just do — and doing it here lands it without the target repository's conventions, in a checkout other work is branching from.
The one exception is the user's notes in ~/Notes. Keeping them current is part of tracking the work, not doing it.
Active Tasks
The user can keep track of three tasks at once. Past that, there's a risk of them getting overwhelmed and losing track. Never start a new task while three or more are active.
To count the active tasks, count the workspaces in herdr workspace list, leaving out this session's own workspace, which isn't a task the user tracks.
Workspaces the user opened themselves are active tasks too. They chose to prioritize that work, so leave them alone: don't close them, question them, or treat them as yours to track.
When three tasks are active, wait. Patience is the job here.
Workflow
- Get the parallelization, which lives at
/tmp/[slugified-title].md. REQUIRED: Use thels-agent:parallelizeskill if there isn't one yet. A task is ready once every task it waits on has finished and fewer than three tasks are active. - Start the watcher (see Watching Workspaces).
- Propose the ready tasks to start, in stage order, without going past three active tasks. Start only the ones the user picks. REQUIRED: Use the
ls-agent:delegateskill once per task. - Report what was started: the task, its project, and the workspace label.
- STOP. Wait for the watcher.
Watching Workspaces
Run scripts/watch-workspaces.sh with the Monitor tool, at its longest timeout. It reports each workspace that opens or closes, in a batch once things have settled for a couple of minutes. Re-arm it each time it expires, for as long as tasks remain. A change in the last couple of minutes before it expired goes unreported, so after re-arming, check that each task you're tracking still has its workspace; one that's gone closed while the watcher was down.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- today First seen · 66 lines · 39 tokens per session scan A 68af380fbdb8
dispatch is a skill published in the GitHub repository LandonSchropp/agent-toolkit (3 stars, last pushed today), licensed MIT. It adds 39 tokens to every session and 1,213 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-10-17.
Other skills, from other repositories
nx-monorepo
Provides comprehensive Nx monorepo management guidance for TypeScript/JavaScript projects. Use when creating Nx workspaces, generating apps/libraries/components, running affected commands, setting up CI/CD, configuring Module Federation, or implementing NestJS backends within Nx.
turborepo-monorepo
Provides comprehensive Turborepo monorepo management guidance for TypeScript/JavaScript projects. Use when creating Turborepo workspaces, configuring turbo.json tasks, setting up Next.js/NestJS apps, managing test pipelines (Vitest/Jest), configuring CI/CD, implementing remote caching, or optimizing build performance…
loom-usage
Meta-orchestration skill for Claude driving loom itself.
cc-skill-project-guidelines-example
Project Guidelines Skill (Example).
capgo-organization-management
Guides Capgo account lookup and organization administration with the CLI, including account whoami, organization list/add/members/set/delete, 2FA enforcement, password policy, API-key expiration and hashed-key policy, and apps per organization. Use when listing or creating Capgo organizations, auditing members' 2FA…
artifact-yylo
Capture and retrieve durable YYLO Ledger artifact Records with intentional profiles, payload modes, provenance, retention, and secret-safe immutable evidence.