mathias/hyperguild
1
0
Fork 0
Code Issues 31 Pull Requests Actions Packages Projects Releases Wiki Activity

docs(close-session): refresh classification gate post-#67 (#62) #69

Merged
mathias merged 1 commits from fix/close-session-classification-refresh into main 2026-06-26 14:28:49 +00:00
Conversation 0 Commits 1 Files Changed 1
+4 -3
1 changed files with 4 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
skills/close-session/SKILL.md
+4 -3
View File
@@ -52,10 +52,11 @@ Persist the session via a **single `capture` call** (the `brain:capture` MCP too
- **`context`** — `{harness: "claudeai-chat", session_ref: <chatid8-or-slug>, fidelity: "live-capture", actor: "mathias", classification: <see gate below>}`.
**THE CLASSIFICATION GATE (read before calling — this is where capture refuses).**
Capture computes an **effective classification = the strictest across EVERY target it touches** (each insight's `wing`, each ticket's `repo`, and every entry in `summary.repos_touched`), then refuses if that effective level is `confidential` and the origin is us-nexus (claude.ai is us-nexus). Server-derived defaults: `hyperguild`/`homelab` → internal; `client-*` → confidential; **anything untagged → confidential (fail-safe)**. There is no populated `classification.yaml` yet, so only these defaults apply.
Capture computes an **effective classification = the strictest across EVERY target it touches** (each insight's `wing`, each ticket's `repo`, and every entry in `summary.repos_touched`), then refuses if that effective level is `confidential` and the origin is us-nexus (claude.ai is us-nexus). Levels come from `classification.yaml` at the brain root (source of truth, #67), with the code defaults as the floor: `hyperguild`/`homelab` → internal; `client-*` → confidential; **anything untagged → confidential (fail-safe)**.
- **Tagged `internal` today** (safe through claude.ai): wings `hyperguild`, `homelab`; repos `brain`, `ai-sessions`, `infra`, `hyperguild`, `homelab`, `tapir`, `agentsquad`, `jepa-fx-risk`, `swedsl`. Treat `classification.yaml` as authoritative — this list is a hint, not gospel.
- Declare `context.classification: "internal"` for normal homelab work.
- **Keep `summary.repos_touched` to genuinely-central, internal-default repos** (e.g. `hyperguild`, `homelab`). Do NOT list untagged repos like `brain` or `ai-sessions` "for completeness" — each one escalates the whole capture to confidential and the gate will refuse via claude.ai. `repos_touched` is a classification INPUT, not free-form metadata. The same caution applies to insight `wing`s and ticket `repo`s: an untagged target escalates the whole call.
- If a session genuinely touched confidential/client material, it cannot be captured through claude.ai at all — note that in the verdict rather than trying to force it.
- `summary.repos_touched`, insight `wing`s, and ticket `repo`s are classification INPUTS, not free-form metadata — every target must resolve `internal` or the whole capture escalates to `confidential` and the gate refuses via claude.ai. Listing the central homelab repos (incl. `brain`/`ai-sessions`) is now fine; they're tagged. The summary always lands in `ai-sessions` (internal), so the summary path itself never escalates.
- If a session genuinely touched **`client-*` or otherwise-untagged** material, it cannot be captured through claude.ai — note that in the verdict rather than trying to force it.
**GATE — dry-run first, then execute.**
1. Call `capture` with `dry_run: true`. It validates the whole payload and returns the would-be receipt + `effective_classification`, writing nothing.