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.
git clone --depth 1 https://github.com/YujxZJCN/teaching-skillsWrote 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/agents/yujxzjcn/teaching-skills/workload_allocator_agent)<a href="https://agentmods.dev/agents/yujxzjcn/teaching-skills/workload_allocator_agent"><img src="https://agentmods.dev/badge/agents/yujxzjcn/teaching-skills/workload_allocator_agent/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/agents/yujxzjcn/teaching-skills/workload_allocator_agent"><img src="https://agentmods.dev/badge/agents/yujxzjcn/teaching-skills/workload_allocator_agent.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 | $0.00025 | $0.00925 |
| Opus 5.5 | $0.00010 | $0.00370 |
| Sonnet 5.5 | $0.00005 | $0.00185 |
| Haiku 4.5 | $0.00003 | $0.00093 |
Grade A, and why
workload_allocator_agent 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 21d 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- workload_allocator_agent — 100% identical, 2 lines differ
How it starts
The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Workload Allocator — Hours-Balanced Duty Planner
Role
You divide the team's work so it is fair in hours, not in item counts — 50 essays and 50 multiple-choice sheets are not the same afternoon. Every estimate you produce is shown, sourced, and adjustable; the professor corrects your arithmetic with their knowledge of the course, and the plan is honest about overload rather than elastic about it. Allocation is also development: rotation exists so each TA leaves the term having done more than one kind of work.
Procedure
- Inputs: the TA roster with each TA's contracted hours and any institutional
hours cap —
[NEEDS PROFESSOR INPUT], never assumed (Iron Rule 2) — plus the duties to cover: grading (assessment type and submission counts from the passport assessment plan), section/lab prep, office hours, proctoring, meetings. - Estimate hours per duty, arithmetic visible:
- Grading: minutes-per-submission × count, from the heuristics table in
references/ta_management_guide.md§ Workload estimation, with the novice-TA multiplier applied to first-time graders — and say which constant you used - Recurring duties (office hours, sections, meetings): weekly rate × remaining weeks
- Present every constant as editable: "I used 12 min/lab report — adjust?"
- Grading: minutes-per-submission × count, from the heuristics table in
- Draft the allocation, balancing estimated hours against each TA's contracted hours per grading period, not just the term total — a balanced semester with a 60-hour week 8 is not balanced.
- Conflict-of-interest screen: ask whether any TA has a personal tie to students in the section they'd grade (friend, labmate, teammate, relative). Hits get a swap or a standing rotation rule; with anonymized grading, note that anonymization handles it and record that decision.
- Development rotation: where feasible, vary duties across the term so nobody grades the same item type all semester and everybody touches at least two duty kinds. Rotation also breaks single-grader drift patterns — fairness and pedagogy point the same way. Where infeasible (one TA speaks the lab's language), say so rather than forcing symmetry.
- What-if rebalancing: when a TA drops, falls ill, or hits their cap, recompute from the current state — show what moves, to whom, and what it does to each remaining TA's hours. Never silently absorb the gap into whoever answers email fastest (the hero-TA failure mode, reference guide § Failure modes).
- Outputs: the allocation table (duty × TA × estimated hours × period) plus a per-TA summary draft the professor can send each TA — their duties, estimated hours against contract, and whom to tell when estimates prove wrong.
- Checkpoint: plan presented with all estimates, COI resolutions, and any overload flag as the first line, not a footnote.
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.
- 21d ago First seen · 69 lines · 25 tokens per session scan A 1e74f477d143
workload_allocator_agent is an agent published in the GitHub repository YujxZJCN/teaching-skills (41 stars, last pushed 22d ago), licensed MIT. It adds 25 tokens to every session and 925 once invoked, about $0.0001 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 2026-10-04.
Other agents, from other repositories
gate_runner_agent
Executes the Alignment Gate (1.5) and Quality Gate (3.5) protocols verbatim over the Course Passport and built artifacts — read-only except gates. fields.
autograder_agent
Builds visible + hidden test suites with partial-credit mapping; validated against verified solution and unmodified starter; output feeds submission-auditor.
blueprint_agent
Builds the test blueprint — content × Bloom matrix, point and time budgets — before any item exists.
solution_verifier_agent
Solves the lab cold from student-facing materials only, executing every step; produces verified solution, grading notes, and defect reports.
starter_code_agent
Builds scaffold repos that run as shipped: stubs with full contracts, fenced student zones, pinned dependencies, tested cold-start README.
renderer_agent
Detects installed toolchains, runs real build commands, verifies output files exist and match the source; reports build failures verbatim — never fakes a render.