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 kennethkhoocy/applied-micro-skills --skill adjudication-sheetsgit clone --depth 1 https://github.com/kennethkhoocy/applied-micro-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/skills/kennethkhoocy/applied-micro-skills/adjudication-sheets)<a href="https://agentmods.dev/skills/kennethkhoocy/applied-micro-skills/adjudication-sheets"><img src="https://agentmods.dev/badge/skills/kennethkhoocy/applied-micro-skills/adjudication-sheets/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/kennethkhoocy/applied-micro-skills/adjudication-sheets"><img src="https://agentmods.dev/badge/skills/kennethkhoocy/applied-micro-skills/adjudication-sheets.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00142 | $0.01112 |
| Opus 5 | $0.00071 | $0.00556 |
| Sonnet 5 | $0.00028 | $0.00222 |
| Haiku 4.5 | $0.00014 | $0.00111 |
Grade A, and why
adjudication-sheets 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 11d 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.
Human Adjudication Sheets from LLM-Pipeline Data
Problem
Adjudication sheets built from pipeline intermediates tend to carry truncated "excerpt" columns (previews made for machine diffing, not human judging). A human asked to rule "does this text show X?" on a fragment produces invalid rulings: the decisive sentence may sit past the cutoff. Separately, multi-line text in CSV cells renders as broken rows/empty cells in Excel, so the labeler reports "there is no data here" even when the column is populated.
Context / Trigger Conditions
- A labeler says the sheet has "no information" — check for embedded newlines in CSV fields first.
- Excerpt lengths cluster at an exact value (all 1,204 / 1,500 / 800 chars) or end with "[…]" — that is a hard cap, not natural length.
- The ruling question is "does the record/text show X?" — the ruling basis must be byte-identical to what the classifier/raters saw, or the ruling grades a different object.
Solution
- Trace the true rating basis from the scoring code, not from intermediates. Find the line where text is passed to the model (e.g.
prompt = ... + text) and reproduce that construction exactly. Verify: stored excerpt == full_text[:cap] (0 mismatches), or cache-key match against the LLM request cache. - Deliver .xlsx, not .csv, for any sheet with multi-line text: wrap_text + frozen header + generous width on evidence columns; data-validation dropdown on the
rulingcolumn with the exact allowed codes (e.g. E_pos/E_neg/unclear, or 0/1/2/unclear). - Handle the Excel cell cap (32,767 chars) explicitly: if full text exceeds it, inline the first ~30k ending with a loud marker ("[CONTINUES — see companion file]") AND write the complete text to a per-case
.txt(header block with case metadata + full text); add atext_filepath column. Write companion files for ALL cases regardless — they are the comfortable reading surface. - Keep the original CSV untouched as the machine-readable artifact; the xlsx is the human ruling surface; harvest rulings from the xlsx.
Distribution:
text_filepaths are project-relative and break the moment the sheet is emailed or copied out of the tree (the labeler asks "where are the attendant/companion files?"). If the labeler is not working inside the project folder, ship ONE zip containing the xlsx and the companion-texts folder side by side, with the instruction: extract together; the file for row X is<texts_folder>\<case_id>.txtnext to the spreadsheet. Require the FILLED xlsx back and harvest from the returned copy, not the original. - Tell the labeler the epistemics: rule from the provided record only; silence = negative; outside research (Google) goes in
ruling_notes, never the ruling — otherwise the labels leak information the classifier could never see.
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.
- 11d ago First seen · 51 lines · 142 tokens per session scan A 1456b3dc441b
adjudication-sheets is a skill published in the GitHub repository kennethkhoocy/applied-micro-skills (28 stars, last pushed 6d ago), licensed MIT. It adds 142 tokens to every session and 1,112 once invoked, about $0.0007 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
stylework-yunxiao-requirement-sync
A workflow for exporting selected requirements from Yunxiao, a project-management service, into a new sheet in an existing DingTalk online spreadsheet. It formats the result into a fixed set of columns and groups it by iteration.
submittal-log-generator
Extract all submittal requirements from specification sections and generate a comprehensive submittal register in Excel. Triggers: 'submittal log', 'extract submittals', 'submittal register'. Requires /spec-splitter output.
schedule-extractor
Extract tabular schedule data from construction drawings — door, window, finish, fixture, panel schedules — and output to Excel. Triggers: 'door schedule', 'extract schedule', 'schedule to Excel', 'panel schedule'.
tag-audit-and-takeoff
Count-based quantity takeoff and tag completeness auditing for construction drawings. Vision + OCR reconciliation, sheet markup, Excel QTO output. Triggers: 'tag audit', 'quantity takeoff', 'QTO', 'count fixtures'.
bid-tabulator
Extract data from subcontractor bid PDFs and produce a comparison spreadsheet. Feeds into /bid-evaluator. Triggers: 'tabulate bids', 'bid comparison', 'compare bids', 'buyout analysis', 'bid tab'.
docs-pipeline-automation
Build repeatable data-to-Docs pipelines from Sheets and Drive sources. Use for automated status reports, template-based document assembly, and scheduled publishing workflows.