tools-doc-ask — IMP FORGE

doc-ask: one command that detects a document's shape, runs the matching sharder (table/transcript/prose), and asks a local Ollama model a question about the resulting opaque shard -- never the raw document. Composes doc-shape + the *-shard family + local-ask's own server-startup/request logic rather than reimplementing any of it (Rule 14).

Tier	tools
Role	unclassified (baselined)
Path	`crates/tools/doc-ask`
Edition	2021
Targets	tools-doc-ask, tools_doc_ask
Public items	6 across 0 modules
Tests	14

What it is for

Pure logic for tools-doc-ask: shard dispatch and default-question construction. I/O (reading the file, starting/calling the model) lives at the edge in main.rs.

Capabilities

`crate root`

Pure logic for tools-doc-ask: shard dispatch and default-question

Item
`fn build_shard_json(text : & str, shape : DocShape, task : & str, allowed_outputs : Vec <String>,) -> String`
`fn default_question(shape : DocShape) -> String`
`fn build_report(shape : DocShape, shard_json : & str) -> String`
`fn example_questions(shape : DocShape) -> Vec <& 'static str>`

`ContentInspection`

Pure logic for tools-doc-ask: shard dispatch and default-question

Item
`pub struct ContentInspection`
`fn inspect_bytes(bytes : & u8) -> ContentInspection`

How to use it

No examples/ target and no doctest in this crate's rustdoc. The tests listed under Verification are the closest executable usage.

Module structure

No public modules: the crate root is its whole surface.

Public surface

`crate root`

Item	What it is
`pub struct ContentInspection`	Content-only anomaly signals -- computed over raw bytes, never reported alongside the actual bytes/content, so this is safe to print even for a file that turns out not to be valid UTF-8 prose at all (the `metadata` route's whole point: cheap, no-LLM triage before deciding whether a document is even the kind of thing this toolchain can shard).
`fn inspect_bytes(bytes : & u8) -> ContentInspection`	Inspect raw bytes for anomalies without assuming they're valid UTF-8 or even text at all -- callers decide what to do with an `is_valid_utf8: false` result rather than this function guessing.
`fn build_shard_json(text : & str, shape : DocShape, task : & str, allowed_outputs : Vec <String>,) -> String`	Build the opaque structural shard for `text` given its already-detected `shape`, and return it as pretty JSON
`fn default_question(shape : DocShape) -> String`	A shape-tailored default question, used when the operator doesn't supply one: always asks the model to explain the JSON fields first (a cheap, checkable grounding test -- a model that gets the field meanings wrong shouldn't be trusted on the substantive questions either), then a handful of real, shape-specific questions worth asking of that shard.
`fn build_report(shape : DocShape, shard_json : & str) -> String`	Build a human-readable, no-LLM summary from an already-built shard's JSON (as produced by `build_shard_json`)
`fn example_questions(shape : DocShape) -> Vec <& 'static str>`	A broader, curated list of example questions per shape, for discovery (`tools-doc-ask questions <path>`) -- richer than `default_question`'s tight 4-question default, covering every signal each shard family actually emits (including the ones added after the first pass: sentiment, POS proportions, embedding clusters).

No pub use re-exports: every item above is declared in this crate.

Boundary

Reaches into domain, infrastructure.

Shares tier tools with 84 other crates: tools-advisory-reach, tools-archive-guard, tools-artifact-scaffold, tools-ask-ai-core, tools-ask-ais, tools-ask-gemini, tools-book, tools-book-report, … (84 total).

_What this crate deliberately does NOT own is a judgment. No committed registry records one for it, so none is stated here._

Where it sits

Tier (ontology)	`tools`
Architectural role (taxonomy)	`unclassified (baselined)`
Location	`crates/tools/doc-ask`
Vocabulary in force (lexicon)	`current`

Tier flow. Which tiers this crate's own edges cross.

flowchart LR
  n_tools["tools"] --> n_domain["domain"]
  n_tools["tools"] --> n_infrastructure["infrastructure"]

Dependencies

Runtime, in this workspace.

Crate	Tier	Optional	Only on
`domain-prose-shard`	domain	no	always
`domain-table-shard`	domain	no	always
`domain-transcript-shard`	domain	no	always
`infrastructure-ai`	infrastructure	no	always
`tools-doc-shape`	tools	no	always
`tools-local-ask`	tools	no	always

Runtime, from outside the workspace.

Crate	Requirement	Features	Optional	Only on
`clap`	`^4`	`derive`	no	always
`serde`	`^1`	`derive`, `derive`	no	always
`serde_json`	`^1`	—	no	always
`tokio`	`^1`	`full`, `macros`, `rt-multi-thread`	no	always

Development, in this workspace.

Crate	Tier	Optional	Only on
`tools-cli-conformance`	tools	no	always

Build. None.

Depended on by. Nothing in this workspace.

Signal flow — what reaches this crate, and what it reaches.

flowchart LR
  SELF["tools-doc-ask"]
  SELF -->|development| n_tools_cli_conformance["tools-cli-conformance"]
  SELF -->|runtime| n_domain_prose_shard["domain-prose-shard"]
  SELF -->|runtime| n_domain_table_shard["domain-table-shard"]
  SELF -->|runtime| n_domain_transcript_shard["domain-transcript-shard"]
  SELF -->|runtime| n_infrastructure_ai["infrastructure-ai"]
  SELF -->|runtime| n_tools_doc_shape["tools-doc-shape"]
  SELF -->|runtime| n_tools_local_ask["tools-local-ask"]
  classDef self fill:#1f883d,stroke:#1f883d,color:#fff;
  class SELF self;

Feature flags

No Cargo features are defined: every capability is unconditional, so no consumer can receive a half-wired crate.

Targets

Kind	Name	Source
bin	`tools-doc-ask`	`src/main.rs`
lib	`tools_doc_ask`	`src/lib.rs`

Error model

No public error type was detected: no public item declares a type named *Error, and no public signature returns one.

Operational characteristics

Property	Evidence
async public surface	none detected
async runtime	yes
database access	none detected
network I/O	none detected
unsafe code	none detected
environment variables	none detected

No unsafe block, unsafe fn, unsafe impl or unsafe trait was found by the parser anywhere in this crate's source.

Configuration

No environment variable is read with a literal name anywhere in this crate. A variable whose key is computed at run time cannot be listed here, and is not claimed to be absent.

No workspace crate depends on this one.

Verification

Kind	Count
Unit tests	14
Integration tests	0
Examples	0
Doctests	0

Evidence by module. How often each public module is named by something executable.

Module	Tests	Examples	Consumers
`crate root`	6	0	0

What the tests establish, by name:

build_report_summarizes_the_real_prose_fixture — src/lib.rs
build_report_summarizes_the_real_tabular_fixture — src/lib.rs
build_report_summarizes_the_real_transcript_fixture — src/lib.rs
build_shard_json_never_leaks_a_real_transcript_speaker_name — src/lib.rs
build_shard_json_produces_valid_json_for_each_shape — src/lib.rs
default_question_differs_per_shape_and_always_asks_to_explain_fields_first — src/lib.rs
example_questions_returns_several_real_questions_per_shape — src/lib.rs
inspect_bytes_counts_null_bytes_and_control_chars — src/lib.rs
inspect_bytes_detects_a_utf8_bom — src/lib.rs
inspect_bytes_detects_invalid_utf8 — src/lib.rs
inspect_bytes_distinguishes_crlf_from_lf_line_endings — src/lib.rs
inspect_bytes_reports_valid_utf8_and_zero_anomalies_for_clean_text — src/lib.rs
inspect_bytes_tracks_the_longest_line — src/lib.rs
cli_definition_conforms — src/main.rs

Documentation coverage

Measure	Documented	Total
Public items with rustdoc	6	6
Public modules with a `//!` block	0	0

pie showData
    title Public items with rustdoc
    "Documented" : 6
    "No rustdoc detected" : 0

Metrics

Metric	Value
Rust source files	2
Source lines	704
Code lines	570
Public API items	6
Public modules	0
Tests	14
Examples	0
Cargo features	0
Direct runtime dependencies	10
Workspace reverse dependencies	0

pie showData
    title Public API by kind
    "function" : 5
    "struct" : 1

pie showData
    title Rust source composition
    "Code" : 570
    "Blank or comment" : 134

Generation

Rendered by tools-corpus corpus readme from repository evidence alone, renderer schema 2, lexicon current. No model, network service or database was consulted. Regenerate with tools-corpus corpus readme --write; verify with --check.