hearing-sheet
Skill5 starsChanged 57 days agoMITView on GitHub
Keep the hearing sheet as data and move it between Markdown, Excel and Google Spreadsheet in both directions — hand one to a customer or partner to fill in, read the answers back, and report what is still unconfirmed. Use for: ヒアリングシートを作って, ヒアリング項目を Excel で, スプレッドシートで顧客に渡したい, 記入してもらったシートを取り込んで, 何が聞けていないか. Not: slides that do the asking (hearing-slides); minutes and email into the record (scalar-deal-intake).
1 other project uses a version of this.
Adopted by
Most recently taken up by
- taiga-shiokawa/claude-skillstheir own version, first seen Sep 2, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- taiga-shiokawa/claude-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository26 other files it gives its agents.
- AGENTS.mdAGENTS.md
- forgeSkill
- analysis-template-creatorSkill
- b2b-account-mapsSkill
- calendar-slidesSkill
- current-state-analysisSkill
- drawio-diagramsSkill
- google-slidesSkill
- google-slides-templateSkill
- hearing-slidesSkill
- image-slotsSkill
- nexus-report-slidesSkill
- pptx-exportSkill
- pricing-strategySkill
- scalar-account-planning-sessionSkill
- scalar-account-planSkill
- scalar-ae-materialsSkill
- scalar-deal-intakeSkill
- scalar-nurture-intakeSkill
- scalar-product-slidesSkill
- scalar-proposal-slidesSkill
- settingsSkill
- slide-qaSkill
- slide-template-creatorSkill
- spreadsheetsSkill
- template-forgeSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 1 is distinctive, and 6 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 18
Read references/hearing-kit.md first — the record, the merge rule, the customer filter and the command list are all there, and this skill …
- Distinctive: line 43
If it is missing, start it from the blank form — and from the product addendum when a product is already in play:
- Very specific: line 55
If a filled-in Markdown sheet already exists from before this skill, init from that file instead of the blank template — the answers come …
- Very specific: line 61
Ask in one batch only what you cannot infer (references/interactive-intake.md §0 and §5):
- Very specific: line 78
Put a customer-facing sheet in 01_顧客提示.
- Very specific: line 93
Run --dry-run first and read the change list.
- Very specific: line 95
A conflict stops the merge: Do not reach for --take to make the message go away: look at both values, decide which is true, and say why in …
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
日本語
Hearing Sheet I/O
Read references/hearing-kit.md first — the record, the merge rule, the
customer filter and the command list are all there, and this skill does not
restate them. This file owns only the workflow and the decisions.
Working directory: the slide-forge root. Run everything with .venv/bin/python.
Boundaries
| Request | Where it goes |
| A hearing sheet in md / xlsx / Google Spreadsheet, in either direction | This skill |
| What is still unconfirmed, and who can answer it | This skill (gaps) |
| Slides that ask the open questions | hearing-slides |
| Minutes, email or a CRM export → the record | scalar-deal-intake |
| Whether the product fits, and which edition | templates/sales/products/<product>.ja.md (a person judges; only the verdict lands here) |
| Gate and BANT judgements | references/scalar/sales-playbook.md |
| A quote or a BOM spreadsheet | spreadsheets |
| The proposal deck | scalar-proposal-slides |
Step 1: Find or start the record
ls accounts/<AE>/<customer>/stages/hearing.jsonIf it is missing, start it from the blank form — and from the product addendum when a product is already in play:
.venv/bin/python scripts/hearing/hearing_sheet.py init \
templates/sales/hearing-sheet.ja.md \
--out accounts/<AE>/<customer>/stages/hearing.json --product scalarDo not start a second record for the same deal. One deal, one
hearing.json, one Google Spreadsheet link.
If a filled-in Markdown sheet already exists from before this skill, init
from that file instead of the blank template — the answers come across, and the
IDs are assigned then.
Step 2: Decide the surface from who is filling it in
Ask in one batch only what you cannot infer (references/interactive-intake.md
§0 and §5):
| # | header | Question | Options |
| 1 | Who fills it in | Who is going to write in this sheet? | We do, from minutes / The customer / A partner / Attendees at an event |
| 2 | Format | Which surface? | Google Spreadsheet (shared link, read back) / Excel (mail it, read the returned file) / Markdown (internal only) |
| 3 | Audience | Is this going outside? | Internal / Handed to the customer (drops the internal columns and sections) |
Decide it yourself when the request already says so. A sheet the customer writes in is a Google Spreadsheet or an Excel file, never Markdown.
.venv/bin/python scripts/hearing/hearing_sheet.py render <json> --format gsheet \
--audience customer --folder <Drive folder URL>Put a customer-facing sheet in 01_顧客提示; an internal one stays out of the
customer folders entirely (sales-playbook.md §8).
Before handing anything over, run the playbook §3 check by eye. The
--audience customer filter removes columns and sections; it cannot tell that a
question is phrased in a way you would not want quoted back.
Step 3: Read what came back
cp <json> <json>.baseline-YYYYMMDD # keep the baseline before rendering
.venv/bin/python scripts/hearing/hearing_sheet.py read <surface|URL> \
--into <json> --baseline <json>.baseline-YYYYMMDD --dry-runRun --dry-run first and read the change list. Then drop it to write.
- A conflict stops the merge. Do not reach for
--taketo make the message go away: look at both values, decide which is true, and say why in the report. - An ID that is not in the record means a row was added or the ID column was edited. Ask before inventing a question to match it.
- Answers arrive as the customer wrote them. Keep their words in 回答; your reading of them belongs in the stage record, not in this cell.
- Confidence is not raised by an answer appearing.
確認済needs the customer to have said it or a document to carry it — if you inferred it, it is推定.
Step 4: Report what is still open
.venv/bin/python scripts/hearing/hearing_sheet.py gaps <json> [--section 4]Report, in Japanese:
- What changed, and where it came from
- What is still
未確認, grouped by section, with who can answer it - Anything answered as
推定— these are whathypothesis-checkwill confirm - Conflicts, with both values and no resolution invented
- What the gaps block downstream — §4.2 and §5 gate the architecture diagram
and the BOM in
scalar-proposal-slides; §14.3 maps each section to the proposal slide it feeds
Then offer hearing-slides for the questions worth asking in person.
Files
| Path | Role |
scripts/hearing/hearing_sheet.py | init / render / read / gaps / validate |
scripts/hearing/model.py | the document shape, the Markdown parser and renderer |
templates/sales/hearing-sheet.ja.md | the blank form (product-neutral) |
templates/sales/products/ | product-fit addenda |
references/hearing-kit.md | shared rules (record, merge, customer filter, storage) |
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

