pbiplint 0.2.3 → 0.2.5

package/NOTICE

@@ -2,6 +2,12 @@ pbiplint itself is licensed under the GNU Affero General Public License,
 version 3 or later; see LICENSE. The MIT license text below applies only
 to the third-party material described here, not to pbiplint.
 
+One file of pbiplint's own is MIT-licensed instead: the skill for AI
+assistants (packages/cli/skill/SKILL.md, skill/SKILL.md in the npm
+package), which pbiplint skill prints and installs, so it can be
+committed to any repository. Its license notice is at the end of the
+file.
+
 pbiplint bundles rule definitions derived from BPARules.json in the
 microsoft/Analysis-Services repository (https://github.com/microsoft/Analysis-Services),
 which is licensed under the MIT License:

package/README.md

@@ -4,8 +4,10 @@ Best-practice linter for Power BI projects. Point it at a PBIP folder, a `.pbip`
 `.SemanticModel` folder, a `.Report` folder, a `definition` folder, or one `.tmdl` file and get
 ranked findings with a link to a fix page for each rule, and, when the input has a report, a
 "Report at a glance" block that says what the report will do when someone opens it. Nothing is
-uploaded: it reads the files you name and writes to your terminal. Node 20.19 or later (or 22.12
-or later).
+uploaded: it reads the files you name and writes to your terminal, and
+[the pbiplint Privacy Promise](https://pbiplint.com/privacy/) says what that covers. Node 20.19 or a
+later 20 release, or 22.12 or later. [The pbiplint CLI](https://pbiplint.com/cli/) page on the site has more:
+where to get it, what it reads, and how to check that it sends nothing.
 
 ```bash
 npx pbiplint path/to/Model.SemanticModel
@@ -13,15 +15,48 @@ npx pbiplint --sample                                       # a bundled project,
 npx pbiplint path/to/model --format sarif --output pbiplint.sarif
 npx pbiplint path/to/model --format markdown
 npx pbiplint rules                                          # every rule with status and severity
+npx pbiplint explain HIDE_FOREIGN_KEYS                      # one rule's guidance, offline (--format json too)
+npx pbiplint path/to/model --quiet                          # the counts, one line per rule
+npx pbiplint path/to/model --rule HIDE_FOREIGN_KEYS         # one rule's findings (repeatable)
+npx pbiplint skill --install claude                         # the skill for AI assistants (--install copilot, codex, gemini; --show)
+npx pbiplint mcp                                            # a local MCP server for an AI assistant to start
 npx pbiplint --help                                         # every option, in one screen
 npx pbiplint --version
 ```
 
+An AI assistant that edits a project can keep its context small: `--quiet` for the counts, one line
+per rule; `--rule <RULE_ID>`, repeatable, for those rules' findings (every rule still runs, the
+summary says how many are shown of how many, and `--fail-on` counts only those); and
+`pbiplint explain <RULE_ID>` for the rule's guidance, from this version, with no network. When there
+are findings, the quiet output's last line names the other two.
+
+`pbiplint skill` prints a skill, in the [Agent Skills](https://agentskills.io/home) format, that tells
+a coding assistant how to use pbiplint. `--install claude`, `copilot`, `codex`, or `gemini` writes it
+where that assistant reads a project's skills, below the current folder; it never replaces a copy
+that differs unless given `--force`, and `--dry-run` writes nothing. `--show` lists where it is and
+whether each copy matches this version.
+
+`pbiplint mcp` is a local MCP server over stdin and stdout, for an assistant that runs local
+servers (Claude Desktop, Claude Code, VS Code, Cursor) to start. Its three tools only read: `lint`
+(what `--format json` prints, or with `quiet` the short summary), `explain_rule`, and `list_rules`.
+It opens no port. See [the CLI page](https://pbiplint.com/cli/#mcp) for each app's setup.
+
 Formats: `text` (default), `json`, `sarif` (for GitHub code scanning and editors), `markdown`.
+The Markdown export writes an email address, an `@name`, a `#123` or `GH-123`, and text
+holding two or more `$` as code, so pasted into a GitHub issue or comment it links no one,
+notifies no one, and renders no math.
 
-Exit codes: `0` no findings at or above `--fail-on`, `1` findings, `2` usage or input error.
+Exit codes: `0` no findings at or above `--fail-on`, `1` findings, `2` a usage error, an input it
+cannot read, or nothing to lint. [What a script can rely on](https://pbiplint.com/cli/#contract)
+states what stdout carries, every exit code, and the JSON documents' fields.
 `--fail-on error` is the default; `--fail-on warning` and `--fail-on info` tighten the gate;
-`--fail-on none` always exits 0.
+`--fail-on none` exits 0 on any run that lints something. A run that can read nothing in the input,
+such as a report saved as a single `report.json` (PBIR-Legacy) with no model beside it, exits `2`
+and says why, rather than passing with no findings.
+
+Pointed at a folder that is not a project but holds one in a folder below it, as a repository
+often does, pbiplint lints that project and names it in a notice. A folder that holds several exits
+`2` and lists them, each with the command that lints it.
 
 ## Configuration
 
@@ -70,8 +105,10 @@ community's published rule files, and are checked against Tabular Editor on the
 the Microsoft rules need statistics only a live model has; they are listed but not run. A column
 whose TMDL names no type, as Power BI Desktop saves most calculated columns, is left out of the
 thirteen ported rules that test a column's type, where Tabular Editor reads the type from the
-column's DAX. Each rule has a page at https://pbiplint.com/rules with what it checks, why, how to
-fix it, and quirks.
+column's DAX. A TMDL line the parser cannot place gets a `PARSE_ISSUE` error on that line rather
+than being dropped silently; that includes a declaration its object does not hold, such as a
+misspelt `columm` under a table or a hierarchy's `level` that lost its tab. Each rule has a page at
+https://pbiplint.com/rules with what it checks, why, how to fix it, and quirks.
 
 The report layer (PBIR) is read beside the model: a `.Report` folder alone is valid input, and with
 the model beside it the two are paired through `definition.pbir` and checked together. The report
@@ -90,4 +127,6 @@ https://github.com/pbiplint/pbiplint.
 
 Copyright (C) 2026 McKinley Consulting. GNU Affero General Public License, version 3 or later; see
 LICENSE. The vendored Microsoft ruleset and PBI Inspector rule metadata are MIT-licensed; see
-NOTICE. The name pbiplint and its logo are trademarks of McKinley Consulting.
+NOTICE. The skill for AI assistants (`skill/SKILL.md`, which `pbiplint skill` prints and installs) is
+MIT-licensed, so it can be committed to any repository; its notice is at the end of the file. The
+name pbiplint and its logo are trademarks of McKinley Consulting.