spreadsheets

Name: spreadsheets
Author: PaulRBerg

117installs3forks

How to install spreadsheets?

paulrberg/agent-skills/spreadsheets

$npx -y skills add PaulRBerg/agent-skills --skill spreadsheets

Installs into the current project.

Summary

Use when CSV, TSV, or Excel (.xlsx) is the primary input/output: inspect, transform, validate, convert, recalc formulas, or create/fix spreadsheets. Do not trigger when tabular data is incidental.

Files · 1

View on GitHub

SKILL.md

1	# Spreadsheets
2
3	Handle tabular data with exact values, minimal diffs, local privacy, atomic writes, and structural validation.
4
5	## Invariants
6
7	1. Keep precision-sensitive amounts as strings and compute with `decimal.Decimal` or DuckDB `DECIMAL(38, 18)`, never
8	binary floats.
9	2. Touch only requested rows, columns, formulas, and formatting. Existing file conventions override house defaults.
10	3. For newly authored text tables, prefer TSV, UTF-8 without BOM, LF, one trailing newline, lowercase `snake_case`
11	headers, ISO dates, `.` decimals, and `-` nulls.
12	4. Read unknown text tables with BOM-tolerant UTF-8; never write a BOM.
13	5. Write in place atomically through a sibling temporary file, validate it, then replace the target.
14	6. Escape external cells beginning with `=`, `+`, or `@`; a bare `-` null is exempt. Formula-prefix cells in trusted
15	authored data are observations, not proof of injection.
16	7. Keep transaction, bank, exchange, and tax data local. Redact samples unless raw rows were explicitly requested.
17
18	## Factual Profiling
19
20	Resolve helper paths from this `SKILL.md`. Profile unknown data before choosing a transformation tool:
21
22	```sh
23	uv run "<skill-dir>/scripts/profile.py" <file> --redact-samples
24	```
25
26	The JSON output has `schema_version: 2`. It reports structural facts, header quality, cardinality/statistics when qsv is
27	available, frequency facts, formula-prefix cells, workbook metadata, and local tool availability. It contains no tool
28	recommendations and does not infer identifiers from uniqueness. Choose the tool from the requested transformation,
29	provenance, output format, and preservation requirements.
30
31	Use `--external-data` only when the cells came from an external or otherwise untrusted source and will be written to a
32	formula-capable consumer. With that flag, formula-prefix cells affect `status`; without it, legitimate formulas such as
33	`=SUM(...)` remain factual observations and do not fail the profile.
34
35	## Tool Routing
36
37	\| Need \| Tool \|
38	\| ------------------------------------------ \| --------------------------------------------------- \|
39	\| Fast structural preview/validation \| `uv run scripts/peek.py <file> --redact-samples` \|
40	\| Factual local quality profile \| `uv run scripts/profile.py <file> --redact-samples` \|
41	\| Counts, stats, frequencies, select, dedupe \| `qsv` \|
42	\| Joins, pivots, aggregation, conversion \| DuckDB with `all_varchar = true` \|
43	\| Exact custom transforms \| `uv run` Python, stdlib `csv`, `decimal.Decimal` \|
44	\| Any `.xlsx`/`.xlsm` input or output \| Read `references/xlsx.md` first \|
45	\| Exact transformation/validation recipes \| Read `references/recipes.md` only when needed \|
46
47	Prefer `qsv --cache-threshold 0` where supported. When qsv stdout must remain TSV, use `-o out.tsv`; stdout otherwise
48	defaults to CSV.
49
50	## Workflow
51
52	1. Inspect with `peek.py --redact-samples`; add `profile.py` when cardinality, formula prefixes, metadata, or available
53	tooling matters. For a no-shape-change edit, save the peek JSON. For intentional row/schema changes, record the
54	expected width and invariants.
55	2. Decide whether formula-prefix cells are dangerous from provenance and output context. Decide the smallest tool that
56	preserves values and formatting. Avoid pandas unless necessary; if used, load every column as strings.
57	3. Apply the transformation atomically. For idempotent appends with legitimate duplicate rows, use multiset difference
58	rather than set deduplication.
59	4. Validate:
60	- unchanged shape: `peek.py --strict --expect-like <before-report>`;
61	- changed shape: `peek.py --strict --expect-columns <n>` plus task-specific counts/keys;
62	- authored house TSV: add `--house`;
63	- formulas: `uv run scripts/recalc.py <file.xlsx>` and require success.
64	5. Report paths, row/column effects, validation, and any workbook features that could not be preserved.
65
66	For human output, lead with `### 📊 Spreadsheet — ✅ updated` only after the write and required validation pass, or
67	`### 📊 Spreadsheet — 🔎 inspected, no files written` for read-only work. On required validation failure, use
68	`### 📊 Spreadsheet — ⛔ not deliverable`. Do not paste private profile JSON or decorate cells, headers, formulas,
69	paths, commands, or diagnostics.
70
71	## prb-finance
72
73	Never hand-edit generated `.pool.tsv`, `.annual.tsv`, or Markdown reports. After source edits, run `just tsv-check`,
74	then `just cli::write-changed`. Cap private output to counts and file references unless raw rows were re

People who liked this also viewed

6 picks

Type

find-skillsHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skSkillsJul 20262.6M27k
grill-meA relentless interview to sharpen a plan or design.SkillsJul 2026633k182k
vercel-react-best-practicesVercel's official collection of agent skillsSkillsJul 2026573k29k
improve-codebase-architectureScan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.SkillsJul 2026522k182k
setup-matt-pocock-skillsConfigure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.SkillsJul 2026454k182k
triageMove issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.SkillsJul 2026424k182k

spreadsheets

117installs3forks

View on GitHub

How to install spreadsheets?

paulrberg/agent-skills/spreadsheets

$npx -y skills add PaulRBerg/agent-skills --skill spreadsheets

Installs into the current project.

Summary

Use when CSV, TSV, or Excel (.xlsx) is the primary input/output: inspect, transform, validate, convert, recalc formulas, or create/fix spreadsheets. Do not trigger when tabular data is incidental.

Files · 1

View on GitHub

SKILL.md

1	# Spreadsheets
2
3	Handle tabular data with exact values, minimal diffs, local privacy, atomic writes, and structural validation.
4
5	## Invariants
6
7	1. Keep precision-sensitive amounts as strings and compute with `decimal.Decimal` or DuckDB `DECIMAL(38, 18)`, never
8	binary floats.
9	2. Touch only requested rows, columns, formulas, and formatting. Existing file conventions override house defaults.
10	3. For newly authored text tables, prefer TSV, UTF-8 without BOM, LF, one trailing newline, lowercase `snake_case`
11	headers, ISO dates, `.` decimals, and `-` nulls.
12	4. Read unknown text tables with BOM-tolerant UTF-8; never write a BOM.
13	5. Write in place atomically through a sibling temporary file, validate it, then replace the target.
14	6. Escape external cells beginning with `=`, `+`, or `@`; a bare `-` null is exempt. Formula-prefix cells in trusted
15	authored data are observations, not proof of injection.
16	7. Keep transaction, bank, exchange, and tax data local. Redact samples unless raw rows were explicitly requested.
17
18	## Factual Profiling
19
20	Resolve helper paths from this `SKILL.md`. Profile unknown data before choosing a transformation tool:
21
22	```sh
23	uv run "<skill-dir>/scripts/profile.py" <file> --redact-samples
24	```
25
26	The JSON output has `schema_version: 2`. It reports structural facts, header quality, cardinality/statistics when qsv is
27	available, frequency facts, formula-prefix cells, workbook metadata, and local tool availability. It contains no tool
28	recommendations and does not infer identifiers from uniqueness. Choose the tool from the requested transformation,
29	provenance, output format, and preservation requirements.
30
31	Use `--external-data` only when the cells came from an external or otherwise untrusted source and will be written to a
32	formula-capable consumer. With that flag, formula-prefix cells affect `status`; without it, legitimate formulas such as
33	`=SUM(...)` remain factual observations and do not fail the profile.
34
35	## Tool Routing
36
37	\| Need \| Tool \|
38	\| ------------------------------------------ \| --------------------------------------------------- \|
39	\| Fast structural preview/validation \| `uv run scripts/peek.py <file> --redact-samples` \|
40	\| Factual local quality profile \| `uv run scripts/profile.py <file> --redact-samples` \|
41	\| Counts, stats, frequencies, select, dedupe \| `qsv` \|
42	\| Joins, pivots, aggregation, conversion \| DuckDB with `all_varchar = true` \|
43	\| Exact custom transforms \| `uv run` Python, stdlib `csv`, `decimal.Decimal` \|
44	\| Any `.xlsx`/`.xlsm` input or output \| Read `references/xlsx.md` first \|
45	\| Exact transformation/validation recipes \| Read `references/recipes.md` only when needed \|
46
47	Prefer `qsv --cache-threshold 0` where supported. When qsv stdout must remain TSV, use `-o out.tsv`; stdout otherwise
48	defaults to CSV.
49
50	## Workflow
51
52	1. Inspect with `peek.py --redact-samples`; add `profile.py` when cardinality, formula prefixes, metadata, or available
53	tooling matters. For a no-shape-change edit, save the peek JSON. For intentional row/schema changes, record the
54	expected width and invariants.
55	2. Decide whether formula-prefix cells are dangerous from provenance and output context. Decide the smallest tool that
56	preserves values and formatting. Avoid pandas unless necessary; if used, load every column as strings.
57	3. Apply the transformation atomically. For idempotent appends with legitimate duplicate rows, use multiset difference
58	rather than set deduplication.
59	4. Validate:
60	- unchanged shape: `peek.py --strict --expect-like <before-report>`;
61	- changed shape: `peek.py --strict --expect-columns <n>` plus task-specific counts/keys;
62	- authored house TSV: add `--house`;
63	- formulas: `uv run scripts/recalc.py <file.xlsx>` and require success.
64	5. Report paths, row/column effects, validation, and any workbook features that could not be preserved.
65
66	For human output, lead with `### 📊 Spreadsheet — ✅ updated` only after the write and required validation pass, or
67	`### 📊 Spreadsheet — 🔎 inspected, no files written` for read-only work. On required validation failure, use
68	`### 📊 Spreadsheet — ⛔ not deliverable`. Do not paste private profile JSON or decorate cells, headers, formulas,
69	paths, commands, or diagnostics.
70
71	## prb-finance
72
73	Never hand-edit generated `.pool.tsv`, `.annual.tsv`, or Markdown reports. After source edits, run `just tsv-check`,
74	then `just cli::write-changed`. Cap private output to counts and file references unless raw rows were re

People who liked this also viewed

6 picks

Type