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 daemonicai/teambuilder --skill teambuilder-apply-dispatch-loopgit clone --depth 1 https://github.com/daemonicai/teambuilderWrote 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/daemonicai/teambuilder/teambuilder-apply-dispatch-loop)<a href="https://agentmods.dev/skills/daemonicai/teambuilder/teambuilder-apply-dispatch-loop"><img src="https://agentmods.dev/badge/skills/daemonicai/teambuilder/teambuilder-apply-dispatch-loop/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/daemonicai/teambuilder/teambuilder-apply-dispatch-loop"><img src="https://agentmods.dev/badge/skills/daemonicai/teambuilder/teambuilder-apply-dispatch-loop.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 | $0.00064 | $0.01232 |
| Opus 5.5 | $0.00026 | $0.00493 |
| Sonnet 5.5 | $0.00013 | $0.00246 |
| Haiku 4.5 | $0.00006 | $0.00123 |
Grade A, and why
teambuilder-apply-dispatch-loop 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 2mo 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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dispatch pending tasks from an OpenSpec change's tasks.md to the right Teambuilder persona (Designer, Programmer, or Tester) in a fresh subagent per task.
Input: The active change name. The caller (typically /opsx:apply) has already selected the change, read context files, and shown current progress — this skill only runs the dispatch loop itself.
Steps
-
Enumerate pending tasks and available personas
Read
openspec/changes/<name>/tasks.mdand collect every- [ ]checklist item in source order, along with the section heading it appears under. Preserve the exact task text.List
.claude/agents/programmer-*.mdand extract the variant suffix from each filename (e.g.,programmer-ios.md→ios). Hold this list for the classification step. If no variant files exist, the list is empty.If there are no pending tasks, exit immediately with a "Nothing to do" message and return control to the caller.
-
For each pending task in source order:
a. Classify by section heading:
- Heading matching
/design|ux|ui/i→ Designer (.claude/agents/designer.md) - Heading matching
/test/i→ Tester (.claude/agents/tester.md) - All other headings, or no heading → Programmer. Pick the variant:
- Check the heading (case-insensitive) for any variant suffix from step 1 as a whole-word substring.
- If exactly one variant matches → use
.claude/agents/programmer-<variant>.md. - If zero or more than one variant matches → use
.claude/agents/programmer.md.
The Architect is expected to structure
tasks.mdheadings to match available personas (see/teambuild:architect). If a heading is ambiguous, the fallback to baseprogrammer.mdis intentional — do not guess.b. Check the persona file exists at
.claude/agents/<persona>.md. If it does not:- Halt the loop immediately.
- Tell the user: "Task N requires the [Persona] persona, but
.claude/agents/<persona>.mddoes not exist. Run/teambuild:<persona>to create it, then re-run/opsx:apply." - Stop here.
c. Dispatch via the Agent tool using the persona's name as
subagent_type. Each persona file in.claude/agents/registers a subagent type via itsname:frontmatter (designer,tester,programmer,programmer-<variant>), so Claude Code loads the persona file as the subagent's system prompt automatically — no read-and-adopt step needed.- Designer task:
subagent_type: "designer" - Tester task:
subagent_type: "tester" - Programmer task:
subagent_type: "programmer"orsubagent_type: "programmer-<variant>"per step 2a.
If the Agent tool rejects the persona name as an unknown subagent type in the current environment, fall back to
subagent_type: "general-purpose"and prepend a line to the prompt: "Read and adopt the persona defined in.claude/agents/<persona>.mdbefore starting."Pass the task text verbatim and file-path pointers only — do NOT embed file contents:
Your task: <task text verbatim from tasks.md> Change artifacts (read these files for context): - Proposal: <absolute path to proposal.md> - Design: <absolute path to design.md> - Specs: <absolute path to specs/ directory> - Tasks: <absolute path to tasks.md> When you have completed the task: 1. Self-verify: confirm the implementation is correct, tests pass (if applicable), and the task claim holds. 2. Mark the task complete in tasks.md by changing its `- [ ]` to `- [x]`. 3. Return a brief summary of what you did. If you cannot complete the task, leave the checkbox unchanged and explain why.d. Verify completion: After the subagent returns, re-read
tasks.mdand confirm the dispatched task is now- [x]. If it is still- [ ]:- Halt the loop immediately.
- Surface the subagent's final message to the user.
- Tell the user which task stalled and wait for guidance.
- Do not advance to the next task.
- Heading matching
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.
- 2mo ago First seen · 83 lines · 64 tokens per session scan A 8c07bbff1a9f
teambuilder-apply-dispatch-loop is a skill published in the GitHub repository daemonicai/teambuilder (11 stars, last pushed 6mo ago), licensed MIT. It adds 64 tokens to every session and 1,232 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
intake
Inspect configured Sources and propose evidence-backed Backlog work.
orchestrate
Toolbox for driving HumanLayer from the CLI - log in, pick a host, launch tasks in their own worktrees, launch sessions, list tasks and their PRs, and copy, move, or upload task artifacts.
devtu-docs-quality
TOP PRIORITY skill — find and immediately fix or remove every piece of wrong, outdated, or redundant information in ToolUniverse docs. Wrong code, broken links, incorrect counts, and overlapping instructions must be fixed or removed — never left in place. Runs five phases: (D) static method scan, (C) static tool-call…
team-lead
Run a goal as a team when you can also do the work yourself: decide what you keep, dispatch the rest, and keep the status board honest. The delta over goal-conductor, which carries the dispatch, patrol and acceptance procedure this one points at. Load it when you are handed a goal, not a task.
shiplog
Recap of everything shipped since the last run - cross-repo PRs, security fixes, star deltas, and X traction, synthesized into a digest article and a ready-to-post shiplog in your voice.
taskmarket-delegate
Delegate work to the TaskMarket agent-worker market - browse open tasks, create tasks with explicit authorization, and track/submit work, so an agent can outsource instead of burning inference on unreliable or low-confidence work.