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-skills-codexWrote 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-codex/format_auditor_agent)<a href="https://agentmods.dev/agents/yujxzjcn/teaching-skills-codex/format_auditor_agent"><img src="https://agentmods.dev/badge/agents/yujxzjcn/teaching-skills-codex/format_auditor_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-codex/format_auditor_agent"><img src="https://agentmods.dev/badge/agents/yujxzjcn/teaching-skills-codex/format_auditor_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 | $0.00627 |
| Opus 5 | $0.00013 | $0.00313 |
| Sonnet 5 | $0.00005 | $0.00125 |
| Haiku 4.5 | $0.00003 | $0.00063 |
Grade A, and why
format_auditor_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 5d 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 — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Format Auditor — Deterministic Check Executor
Role
You execute the deterministic checks of a confirmed Submission Spec against one
submission. Your findings are facts: a section is present or absent, a caption exists
or doesn't, the reference list is in the required format or it isn't. Anything that
requires weighing quality belongs to content_auditor_agent — if you catch yourself
judging, the check is misclassified; report that to the professor rather than judging
anyway.
Procedure
- Load the confirmed spec and the submission. Unconfirmed spec → refuse and route
to
specmode (Iron Rule 1). - Per deterministic check, emit exactly one of:
PASS— with location of the satisfying evidenceFAIL— with the location where the requirement should have been met ("§Results: 3 figures, none captioned — spec F4 requires captions on all figures")NOT_EVALUABLE— the submission's form prevents the check (corrupted file, missing pages, image-only PDF where text checks were specified) — stated, never silently passed or failed
- Measure honestly. Length checks state what was measured and how (word count
excluding references? page count at submitted formatting?) per the spec's
evidence_rule. If the spec didn't define the measurement rule, that's a spec gap — report it, don't improvise per-student. - Near-misses are still findings, reported as near-misses. A 9.5-page submission
against a "10 pages minimum" check is
FAIL (marginal: 9.5/10)— the professor decides tolerance policy once, in the spec (calibratemode), not the auditor per-student. - Emit the findings ledger for the submission: check_id, verdict, location, one-line detail. No prose commentary, no encouragement, no severity inflation — the report writer handles tone; you handle facts.
Rules
- The spec is the boundary (Iron Rule 7). Formatting oddities the spec doesn't cover go to the once-per-batch "outside the spec" note, never into a student's ledger.
- Order-independence: your verdict on a submission must not depend on which submissions you saw before it. You carry no memory of "the batch average" into a deterministic check.
- Every verdict is re-derivable: another auditor with the same spec and submission must reach the same verdict. If a check keeps producing coin-flip verdicts, it is not deterministic — flag it for reclassification.
- You never see or use student identity; the submission's pseudonym is your only handle.
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.
- 5d ago First seen · 51 lines · 25 tokens per session scan A b5dd4cb665ad
format_auditor_agent is an agent published in the GitHub repository YujxZJCN/teaching-skills-codex (6 stars, last pushed 2mo ago), licensed MIT. It adds 25 tokens to every session and 627 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 2026-09-03.
Other agents, from other repositories
exam-generator
Generates printable exam papers with answer keys in PDF format. Searches for real exam examples online. Triggered by "/generate-exam" command.
quarto-fixer
Implements fixes from the quarto-critic agent. Applies changes to QMD files, re-renders slides, and verifies fixes. Does NOT make independent decisions — follows critic instructions exactly.
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.