55a80ce6e3
- Add ai-core/ — Chief, Scout, Scribe OpenClaw agents + shared knowledge base - Consolidate research/reference docs and the sns.md brand foundation under ai-core/knowledge/ - Repoint every sns.md reference (business branding, divisions, root README) to ai-core/knowledge/sns.md - Update root + ai-core READMEs to reflect the new structure
1.6 KiB
1.6 KiB
KNOWLEDGE — Connected Apps (Scout's read-only allow-list)
This is the only set of external apps I may sign into. If an app isn't listed here as active, I don't touch it — I hand the task back to Chief.
What this file is (and its limits)
- It declares what is allowed — it does not by itself enforce anything. Real enforcement is the token's read-only scope + the gateway secret store + network egress rules. This is the policy layer of defense-in-depth; it only works paired with those technical controls.
- Credentials are never stored here. Each row points to a secret name in the gateway secret store; the value lives only there — never in this file, my workspace, briefs, the research log, or git.
- Adding or changing a row requires Sam. Least privilege: read-only, one app at a time.
Allow-list
| App | Facts I gather | Access | Scope | Secret ref (name only) | Status |
|---|---|---|---|---|---|
| (none connected yet) | — | — | — | — | — |
Rules
- Only rows marked active are permitted. Anything else → stop and hand back to Chief.
- Read / list / export endpoints only — never create, update, delete, or post.
- If a provided token appears to have more than read access, I refuse to use it and flag Sam.
- Offboarding an app = revoke the token at the provider and remove/deactivate its row here.
- I log what I accessed (source + date) in
../projects/research-log.md— never the secret.
Update only with Sam's approval, alongside creating/revoking the matching RO token.