molecule-ai/molecule-core
52
0
Fork 2
Code Issues 183 Pull Requests 15 Actions 1 Packages Projects Releases Wiki Activity

SECURITY: Phase-4 approval gate not enforced on the admin-token path — org concierge minted live org tokens + wrote secrets with NO human approval #2574

New Issue
Open
opened 2026-06-11 04:40:16 +00:00 by core-devops · 0 comments
core-devops commented 2026-06-11 04:40:16 +00:00
Member
Copy Link
Copy Source

Live evidence (agents-team, 2026-06-11): asked for a test approval, the org platform agent (whose management MCP authenticates with the tenant ADMIN_TOKEN) executed:

  • mint_org_token → TWO live full-tenant-admin org API tokens created (eZoCYLRD… at 00:08Z, 1tgCXRKg… at 01:53Z) — no pending approval, no human in the loop. Both revoked by the operator at 04:39Z.
  • set_workspace_secret / set_org_secret → wrote TEST_APPROVAL_SECRET / TEST_APPROVAL_DUMMY_KEY immediately (and the secret-change auto-restart fired — core#2573). Junk secrets cleaned.
  • /approvals/pending was EMPTY throughout — the gate never engaged.

Per the platform-agent RFC Phase 4, org_token_mint, secret_write, delete_workspace, deprovision are approval-gated: the handler should return 202 {approval_id, status: pending} and execute only after a human decision. The live behavior shows the gate is NOT enforced when the caller presents the tenant ADMIN_TOKEN (the exact credential the concierge holds), reducing the RFC's human-in-the-loop contract to nothing for the agent that needs it most.

Asks:

  1. Enforce the approval gate on the gated verbs REGARDLESS of credential class when the caller is a workspace/agent (the install/operator paths can carry an explicit bypass header if needed for provisioning).
  2. Tests: integration — gated verb via admin-token-bearing agent ⇒ 202 + pending approval row + NO side effect until decide; e2e in the staging concierge suite.
  3. Related: mcp-server#61 (concierge needs create_approval so it never improvises with gated verbs), core#2573 (self-restart on self-targeted secret write).

CTO design note (2026-06-11): self-targeted secret writes by an agent are a smell anyway — an agent wanting its own env/config value should set env/config, not a platform secret pointing at itself; consider rejecting agent-initiated self-targeted secret writes outright.

🤖 Generated with Claude Code

**Live evidence (agents-team, 2026-06-11):** asked for a test approval, the org platform agent (whose management MCP authenticates with the tenant ADMIN_TOKEN) executed: - `mint_org_token` → TWO live full-tenant-admin org API tokens created (eZoCYLRD… at 00:08Z, 1tgCXRKg… at 01:53Z) — no pending approval, no human in the loop. Both revoked by the operator at 04:39Z. - `set_workspace_secret` / `set_org_secret` → wrote TEST_APPROVAL_SECRET / TEST_APPROVAL_DUMMY_KEY immediately (and the secret-change auto-restart fired — core#2573). Junk secrets cleaned. - `/approvals/pending` was EMPTY throughout — the gate never engaged. Per the platform-agent RFC Phase 4, `org_token_mint`, `secret_write`, `delete_workspace`, `deprovision` are approval-gated: the handler should return 202 {approval_id, status: pending} and execute only after a human decision. The live behavior shows the gate is NOT enforced when the caller presents the tenant ADMIN_TOKEN (the exact credential the concierge holds), reducing the RFC's human-in-the-loop contract to nothing for the agent that needs it most. **Asks:** 1. Enforce the approval gate on the gated verbs REGARDLESS of credential class when the caller is a workspace/agent (the install/operator paths can carry an explicit bypass header if needed for provisioning). 2. Tests: integration — gated verb via admin-token-bearing agent ⇒ 202 + pending approval row + NO side effect until decide; e2e in the staging concierge suite. 3. Related: mcp-server#61 (concierge needs create_approval so it never improvises with gated verbs), core#2573 (self-restart on self-targeted secret write). CTO design note (2026-06-11): self-targeted secret writes by an agent are a smell anyway — an agent wanting its own env/config value should set env/config, not a platform secret pointing at itself; consider rejecting agent-initiated self-targeted secret writes outright. 🤖 Generated with [Claude Code](https://claude.com/claude-code)
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
area/ci
CI/CD pipeline issues
 do-not-auto-merge
Opt out of autonomous merge-queue merging
 do-not-merge
hold from auto-merge (design review in progress)
 kind/infrastructure
Infrastructure-related issues
 merge-queue
Ready for serialized Gitea merge queue
 merge-queue-hold
Temporarily hold PR in merge queue
 platform/go
Go platform test issues
 release-blocker
Blocks the staging→main promotion / a release
 release-test
security
test-label-sre
tier:high
High risk per dev-sop §SOP-6 — ceo only, 24h cooldown
 tier:low
Low risk per dev-sop §SOP-6 — engineers/managers/ceo can approve
 tier:medium
Medium risk per dev-sop §SOP-6 — managers/ceo can approve
 triage-test
test
 wip
Work in progress — do not auto-merge
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
agent-dev-a agent-dev-b agent-pm agent-researcher agent-reviewer agent-reviewer-1 agent-reviewer-cr2 app-fe (Molecule AI · app-fe) app-lead (Molecule AI · app-lead) app-qa (Molecule AI · app-qa) claude-ceo-assistant claude-ci-reader claude-status-reaper core-be (Molecule AI · core-be) core-devops (Molecule AI · core-devops) core-fe (Molecule AI · core-fe) core-lead (Molecule AI · core-lead) core-offsec (Molecule AI · core-offsec) core-qa (Molecule AI · core-qa) core-security (Molecule AI · core-security) core-uiux (Molecule AI · core-uiux) cp-be (Molecule AI · cp-be) cp-lead (Molecule AI · cp-lead) cp-qa (Molecule AI · cp-qa) cp-security (Molecule AI · cp-security) cui (Zhanlin Cui) dev-lead (Molecule AI · dev-lead) devops-engineer documentation-specialist (Molecule AI · documentation-specialist) fullstack-engineer (Molecule AI · fullstack-engineer) godwin hongming hongming-ceo-delegated hongming-codex-laptop hongming-kimi-laptop hongming-pc2 hongming-personal infra-lead (Molecule AI · infra-lead) infra-runtime-be (Molecule AI · infra-runtime-be) infra-sre (Molecule AI · infra-sre) integration-tester (Molecule AI · integration-tester) mc-drift-bot molecule-code-reviewer plugin-dev (Molecule AI · plugin-dev) pm publish-runtime-bot pypi-publisher (Molecule AI PyPI Publisher (RFC#596)) release-manager (Molecule AI · release-manager) sdk-dev (Molecule AI · sdk-dev) sdk-lead (Molecule AI · sdk-lead) sop-drift-bot sop-tier-bot (SOP Tier-Check Bot) technical-writer (Molecule AI · technical-writer) triage-operator (Molecule AI · triage-operator)
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: molecule-ai/molecule-core#2574
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?