Files
hyperguild/ingestion/internal/capture/ports.go
T
mathiasandClaude Opus 5 6e5a9cfce3
CI / Mirror to GitHub (push) Successful in 4s
CI / Lint / Test / Vet (push) Successful in 35s
fix(brain): unwrap wikilinks that name no entry at the write boundary
Every brain write accepted [[targets]] that resolve to nothing — skill
names, repo names, agent-local memory-file slugs — and each one became a
permanent [high] broken-link finding. The brain-audit digest grew 19 -> 63
high findings between 2026-08-26 and 2026-09-16; 53 of the 63 were exactly
this, and none are fixable by brain-gardener's BrokenLinkFixer, which only
repoints edit-distance-1 renames. There is no target to repoint to.

StripUnresolvedLinks resolves each link against knowledge/ + wiki/ using
brain-gardener's own resolver semantics (bare slug or wing/hall/slug path
form, code spans excluded) and unwraps the misses to plain text, reporting
them as Ref.Unlinked / the MCP `unlinked` field.

Unwrap rather than reject: a refused write on claudewatcher's unattended
ingest path is a lost learning, and the author's words are worth keeping
even when the link is not. An empty corpus is a no-op, so a misconfigured
brainDir cannot strip every link in a note.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-09-16 23:53:42 +02:00

126 lines
4.5 KiB
Go

package capture
import (
"context"
"time"
"github.com/mathiasbq/hyperguild/ingestion/internal/classification"
)
// Ref is the read-after-write handle returned by a brain write/update —
// the #45 contract. ContentHash lets the caller confirm what landed
// without a re-query; for an Update, Superseded is true.
type Ref struct {
ID string
Path string
ContentHash string
Superseded bool
// Unlinked names the [[wikilink]] targets the store unwrapped to plain
// text because they resolve to no entry. Reported rather than rejected:
// see brain.StripUnresolvedLinks for why the write still lands.
Unlinked []string
}
// StoredNote is a brain note fetched by Get: the read-after-write
// confirmation primitive (a direct fetch, never a semantic query).
type StoredNote struct {
ID string
Path string
ContentHash string
Frontmatter map[string]string
Body string
}
// Note is the brain-write payload. It carries both the wing/hall taxonomy
// and the legacy type/domain fields so a single BrainStore serves both
// capture insights and the existing MCP brain_write surface. Reason is
// the supersede rationale, used only by Update.
type Note struct {
Content string
Filename string
Wing string
Hall string
Type string
Domain string
Reason string
}
// BrainStore is the brain persistence port — the shared implementation of
// the #45 write/update/get verbs that both the MCP handlers and capture
// call, so there is one implementation, not two. The read-after-write +
// staleness discipline lives behind this interface so no caller carries
// the rule.
type BrainStore interface {
Write(ctx context.Context, n Note) (Ref, error)
Update(ctx context.Context, slug string, n Note) (Ref, error)
Get(ctx context.Context, id string) (StoredNote, error)
}
// IssueRef identifies a ticket touched by the tracker.
type IssueRef struct {
Repo string
Number int
URL string
}
// IssueTracker is the Gitea ticket port. The implementation (#52) always
// scopes to owner "mathias"; the port deliberately omits owner.
type IssueTracker interface {
CreateIssue(ctx context.Context, repo, title, body string) (IssueRef, error)
// CloseIssue closes an issue, optionally posting a closing comment
// first (empty comment ⇒ close only).
CloseIssue(ctx context.Context, repo string, number int, comment string) (IssueRef, error)
CommentIssue(ctx context.Context, repo string, number int, body string) (IssueRef, error)
}
// ClassificationPolicy derives a target's sensitivity (model C). The
// "stricter wins" combination of declared vs derived is use-case policy
// and lives in the service, so the port stays minimal. Satisfied by
// classification.Config (#50).
type ClassificationPolicy interface {
Derive(target classification.Target) classification.Level
}
// AuditEntry is the request-level audit record (I5): who/what captured
// what, when, via which principal. SecurityEvents carries anomalies such
// as a caller under-declaring sensitivity relative to the target floor.
type AuditEntry struct {
Timestamp time.Time
Principal string
Actor string
Harness string
SessionRef string
EffectiveClassification string
Items []string
SecurityEvents []string
}
// AuditOutcome is how a capture's audit record was (or will be) persisted.
type AuditOutcome int
const (
// AuditCentral means the record goes to the central sink (loki).
AuditCentral AuditOutcome = iota
// AuditBuffered means the central sink was unreachable and the record
// is written to a durable local buffer for later reconciliation
// (internal/public tier only).
AuditBuffered
)
// AuditSink is the two-phase, classification-aware audit port (I5, §4.4).
//
// Reserve runs BEFORE any write and decides whether the capture can be
// audited at its effective classification: it returns the outcome to use,
// or an error to refuse the capture before anything is written
// (confidential + central sink down → refuse; the all-tiers floor when
// nothing can record → refuse). Record runs AFTER the writes and persists
// the final entry per the reserved outcome.
//
// Splitting reserve from record is what lets "confidential + sink-down →
// refuse before any write" be literally true while the record itself
// (which lists what landed) is necessarily written afterwards.
type AuditSink interface {
Reserve(ctx context.Context, level classification.Level) (AuditOutcome, error)
Record(ctx context.Context, e AuditEntry, outcome AuditOutcome) error
}