Janady13/missionmatch / internal/platform/extraction/extraction.go

Source identity

Field	Recorded value
Source type	GitHub source
Original record	Janady13/missionmatch / internal/platform/extraction/extraction.go
Date	Not specified in source
Repository	Janady13/missionmatch
Revision	720d47fae476b0d38477ae4401697d372b1ea2bc
Branch	feature/opportunity-to-outcome
Path or archive member	internal/platform/extraction/extraction.go
Original SHA-256	d52e2925b3f6de0bb4cc02a40a3542abe78047dc42e0035d7e585793b6a9e44d
Git blob SHA	337a21a09d8100724c5fc134577abfad164f3788
Reading-copy SHA-256	d52e2925b3f6de0bb4cc02a40a3542abe78047dc42e0035d7e585793b6a9e44d

Original source ↗

Every recorded occurrence

These 35 records contain identical original source bytes. Their paths and revisions remain part of the archive.

Original record	Branch or archive	Revision / identifier	Original location
Janady13/missionmatch / internal/platform/extraction/extraction.go	feature/opportunity-to-outcome	720d47fae476b0d38477ae4401697d372b1ea2bc	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	per-tenant-routing	9bde0f9a74b89d01980fa658e5253acda7528d0c	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	collision-deconfliction	4ee5a09972608ed89f8377c6665542d5f450f0a6	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	ws6-retrieval	6c56e29231ec05735927fa265c97d517499dfeaf	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	workstream-a-shared-core	e0de03ac3a1ec826bd15d0e048e3b02baa23eeff	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	m5-price-to-win	41000ad30fff4129fdaf5ec84e15ba14c27d4890	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	workstream-c-slatematch	5cd60c918301b46ecd6ed3e03e20c395da0062e2	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	ws1-ontology-sync	b596f449a32a1016fbd403c36505cc00a1ac9ee6	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	master	1e9249840a4302ec82e123b9c9cb6f75badea3e4	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	ws8-teaching-loop	c823f0fb6bb2aba0f39001eb6de96f4000d9321f	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	w3-ai-staffing-recommendations	43d56387270ef4da4cabd794d97ca6a49909bc7d	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	mag-substrate	b82c7100a492dec67dfd1eda6b3f53def859be35	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	ws3-clearance	4b8e7ce594c068621867f9c9deed0ec64ae6c553	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	workstream-d-command-center	255e7d1d420abf5e277d0cd8c9b14f6f72b6acb9	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	phase2-eval-fairness	d361af254701975fe239927656ababb56374655e	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	workstream-e-dataflow-deploy	433815a042f9741411e60fa553fe72143e9b10db	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	ruthless-debug-sweep	b7d742f8f032984be3817efffa3eb8b13b651c1f	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	time-to-fill-survival	b66e1e505863d4f17741a4b9ede73509e45bd6c6	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	ai-intelligence-layer	5d5ddff70393a197c5f54a45280bb5dd42b24755	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	graphrag-foundation	b0ddeb9f633fd138d09309d1a17b7d13bc3a67f7	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	workstream-a-opportunity-intelligence	1675031db74316e100782d063e9eda1a597fe3a1	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	deploy-and-ops	bdacd9d0b4d5998d25fc74bf4007f4c3ec538c7a	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	ws9-bench-ingestion	f585a9ef32fa554663ce1fba9a414a54b6e49cf0	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	ws2-alias-relationship	e4dada4601c6eb101830f9d223a9655dfa120616	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	presentation-readiness	eb574acaa29d71ce4d0f852572e87c275c79fdb7	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	vendor-intelligence	fe9796a19c261951d3f2565206c0c1b5ac294465	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	workstream-b-screenmatch	09015496d565a172bcaa15cfe9a8e4b072b13df0	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	ws7-federal-signal	6a9f6ce9cfb13345ffb47770d4960e7c2d9c17c2	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	workstream-e-commercial-stubs	4437da660539284a400ee380b9b427a300c369ee	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	ai-brain-live	aac32fc49c878cc5bca099459307611562d0bac1	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	workstream-d-missionmatch	3be46ba89edb737182075a2b320509b42dc916f8	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	phase2-connectors	e038d4c687aec16f8bd9bc4d907eea14d91f8704	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	workstream-b-sam-contract-awards	d5101cceb98e70000cebb31d3e7ac73d747d3a6d	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	feature/aioe-upgrade	2178e7298d0e5d9af83afa12402fd587f0ba6ed8	Open original
Janady13/missionmatch / internal/platform/extraction/extraction.go	capture-readiness	bae6af7cb2997e819174a9e9cfff5e16067a1975	Open original

Recovered source transcript

Download complete source text ↓66 lines · 3.8 KB

// Package extraction turns plain resume or job description TEXT into a flat set of proposed structured
// fields (the RawExtraction). It is the server side half of the W1 Mission Check "model plus zod"
// step: the browser does the keyless file to text conversion, and this package does the structured
// extraction behind auth so no provider key ever reaches the browser bundle.
//
// It is deliberately domain light: it emits stable DB code strings (for example "ts_sci", "secret")
// and raw skill, certification, and agency mentions, but it does NOT import the scoring engine. The
// matching module is the authority that maps a RawExtraction onto the engine input (the normalizer),
// so the extractor stays a pure, testable text parser with no scoring or database dependency.
//
// This pass ships only the deterministic extractor, which is production safe for the closed vocabulary
// fields (clearance, certifications, agencies) and a curated skills gazetteer. A future provider backed
// extractor (self hosted vLLM for CUI in production) can implement the same interface behind the port
// without touching the scoring path.
package extraction

import (
	"context"
	"fmt"

	"github.com/missionworkforce/missionmatch/internal/platform/config"
)

// RawExtraction is a flat set of PROPOSED field values parsed from a resume or a job description.
// Clearance fields carry stable DB codes ("ts_sci", "secret", "full_scope", "ci"), empty when not
// found. Every other string is a raw mention. The matching module clamps and maps these onto the
// engine input; the extractor never asserts a value is verified.
type RawExtraction struct {
	FullName        string   // resume only, used only to prefill the optional Save Candidate form
	ClearanceLevel  string   // DB code: "confidential" | "secret" | "top_secret" | "ts_sci" | "" (none found)
	ClearancePoly   string   // DB code: "ci" | "full_scope" | "" (none found)
	ClearanceActive bool     // the text asserts a current clearance and does not say expired or inactive
	Skills          []string // raw skill mentions resolved against the gazetteer
	Certifications  []string // raw certification mentions resolved against the gazetteer
	Agencies        []string // raw customer or agency mentions resolved against the gazetteer
	YearsExperience int      // largest plausible years figure found
	Location        string   // a detected work location, empty when none found
}

// HasSignal reports whether enough was parsed to score against. An empty or garbage upload that yields
// no skills, no clearance, no years, and no certifications is rejected by the caller rather than scored.
func (rx RawExtraction) HasSignal() bool {
	return len(rx.Skills) > 0 || rx.ClearanceLevel != "" || rx.YearsExperience > 0 || len(rx.Certifications) > 0
}

// Extractor parses resume or job description text into a RawExtraction. Implementations MUST be
// deterministic for a given input so Mission Check is reproducible (the AEDT mandate).
type Extractor interface {
	// ExtractResume parses resume text into proposed candidate fields.
	ExtractResume(ctx context.Context, text string) (RawExtraction, error)
	// ExtractRequirement parses a pasted job description into proposed requirement fields.
	ExtractRequirement(ctx context.Context, text string) (RawExtraction, error)
}

// New selects the extractor from config. This pass supports only the deterministic extractor; an empty
// provider also selects it so a default deployment works with no extra configuration. A provider that
// is named but not yet implemented is a hard error so a misconfiguration never falls back silently.
func New(cfg config.Extraction) (Extractor, error) {
	switch cfg.Provider {
	case config.ExtractorDeterministic, "":
		return NewDeterministic(), nil
	default:
		return nil, fmt.Errorf("extraction: provider %q is not implemented in this build", cfg.Provider)
	}
}