## Why
An AI agent autonomously deleted a live task (FN-8609) via
`fn_task_delete`. Investigating that led to two audits, which found the
gate machinery is well-built and its *enforcement* well-tested — but the
*decision boundary* was unguarded and the tools that matter never
reached the gate at all.
The headline defect: all ~76 tools in `packages/cli/src/extension.ts`
are delivered through `hostExtensionPaths` → `DefaultResourceLoader`,
which never passes through `wrapToolsWithActionGate`. `fn_task_delete`
*was* classified as gated in `gating-classifications.ts` — that
classification was dead code. Roughly 15 destructive tools were ungated
even for an operator who had locked their policy all the way down.
Second: `POST /api/approvals/:id/decision` took the approver identity
from the request body, validated for shape only, with no check that the
decider was human or differed from the requester. Combined with a bash
tool whose command string was never inspected (only `params.cwd`), an
agent could read the daemon token from `~/.fusion/settings.json`, find
its own pending request, and approve it.
## Scope
This does **not** tighten defaults. The `unrestricted` preset is
byte-for-byte unchanged — `git diff` on `agent-permission-policy.ts` is
empty — and regression tests assert that an out-of-the-box install
behaves exactly as before. The bug was never "the default is too
permissive"; it was "strict policy doesn't enforce." This makes turning
security up actually work.
The one deliberate exception: the containment that stops an agent
escalating its *own* privileges (reading the daemon token / credentials,
calling the approvals API to self-approve) applies at every preset
including `unrestricted`. That is a privilege-escalation boundary rather
than a permission preference — if it only engaged under strict policy it
would not have prevented the incident that prompted this.
## What changed
8 bisectable commits:
- **Approval lifecycle** — self-approval blocked via server-derived
deciders; same-verdict replay 409s; decide re-reads and re-validates
inside the transaction; expiry TTLs; `markCompleted` ownership check;
session identity registry in core.
- **Engine gates enforce for real** — unclassified tools resolve to a
policy-governed category instead of hardcoded `allow`; missing-policy
fail-open closed; bash containment floor + exact-command approval
binding.
- **Dashboard decision routes** — stop trusting client-supplied actors
(decision, bypass-review, worktrunk → 403 on forged actors).
- **`fn serve` authenticated by default** — auto-mints a token following
the existing `fn dashboard` precedent; `--no-auth` opts out.
- **Sibling entry points closed** — user-sourced hard-cancel moves, ACP
execute-once approvals, plugin task-store gating.
- **pi-extension principal resolution** — the extension resolves the
acting principal and can withhold or policy-gate the previously ungated
destructive tools.
- **Root-cause bonus fix** — `findLatestByDedupeKey` was broken in
PostgreSQL backend mode (already-parsed jsonb fed through a string-only
parser), so approved-grant redemption **never matched in production**,
minting duplicate requests. This explains the live DB state of 17
approved / 0 completed. *(Also cherry-picked to `main` as `a9b30013bb`,
since it is an active production defect on its own.)*
- **Review follow-ups** (`627f1b1fa8`) — operator-configured
provisioning privilege and a configurable grant TTL; see below.
## Review follow-ups
**Provisioning privilege is operator-configured, not role-derived.**
`isCallerPrivileged` had gone from `caller.reportsTo == null` (every
top-level agent privileged — permanent escalation by creating a
manager-less agent) to `caller.role === "ceo"`, which swapped an
implicit rule for a magic string: any agent config can claim that role,
while an operator who genuinely wants a privileged agent had no
supported way to say so. Privilege now derives solely from
`agentProvisioning.trustedAgentIds` / `trustedRoles` and fails closed
when settings are unresolvable.
It is also no longer forwarded to `resolveAgentProvisioningPolicy` as
`isPrivileged`, because that flag short-circuits ahead of
`alwaysApproveDelete` — a trusted caller was bypassing delete approval
entirely. The policy applies the same trusted rules itself, in the right
order. The function now governs only the org-chart escape hatch (acting
outside your own direct reports).
**Grant TTL defaults to 1 hour and is configurable.** Approval →
redemption is not instantaneous: an operator approving from their phone,
an engine restart, a queued lane, or a task waiting on a worktree all
routinely exceeded 15 minutes, after which the grant expired and the
agent silently re-requested. One hour remains far short of the
"redeemable forever" hazard the TTL exists to bound. Override via
`FUSION_APPROVAL_GRANT_TTL_MS` or `configureApprovalRequestTtls()`;
invalid overrides are ignored rather than widening the window to
infinity or collapsing it to zero.
## Behavior changes requiring operator review before rollout
1. `fn serve` requires a bearer token by default (`--no-auth` opts out);
unauthenticated clients get 401.
2. Agents can no longer run withheld destructive tools
(`fn_task_delete`, `fn_task_bypass_review`,
mission/milestone/slice/feature/workflow deletes, `experiment_finalize`,
`skills_install`). Operators keep them via CLI/dashboard. **This is the
incident fix.**
3. Agents get provisioning privilege only when the operator lists them
in `agentProvisioning.trustedAgentIds` / `trustedRoles`; the
provisioning gate is now live in production. Previously-implicit
privilege (top-level position, or a `ceo` role) no longer grants
anything on its own.
4. Decision replay 409s (was 200); pending approvals expire after 24h,
approved grants after 1h (configurable); bash approvals bind per exact
command.
5. Forged/body actors on decision, bypass-review, worktrunk routes →
403; `archive-all-done` requires `{confirm:true}` (external scripts
affected).
6. `fn_secret_get` approvals grant exactly one reveal (previously
granted nothing and looped forever); ACP approvals are execute-once
(previously infinite reuse).
7. Bash containment denies token/credential/approvals-API commands in
all agent sessions at every preset.
## Verification
Independently re-run against the branch, not just self-reported:
- 5 typechecks (core, engine, cli, dashboard `tsconfig.json` +
`tsconfig.app.json`) — clean
- `pnpm lint` — clean
- `pnpm test:gate` — 379 passed
- `pnpm build --force` — green (a plain `pnpm build` skips packages as
unchanged and does **not** compile the branch)
- `pnpm check:changesets` — clean
- ~650 file-scoped tests including new negative-path suites for the
decision boundary, which previously had **zero** test coverage
`packages/engine/src/__tests__/plugin-runner.test.ts` fails 56/80 —
**verified pre-existing**, reproducing identically at base commit
`93a403af67` on `main`. Not in the merge gate.
### A mutation check that failed to fail
Worth recording, because it nearly shipped an untested security fix. The
first mutation check on the provisioning change reintroduced the `ceo`
hardcode and **all 17 tests still passed** — the tests asserted through
the policy path, which can no longer observe `isCallerPrivileged` at
all, precisely because `isPrivileged` is no longer forwarded there.
Org-chart cases that do exercise the function were added; the hardcode
now fails exactly 1 of 19, and restoring is green. A green mutation run
is only meaningful if the test can actually see the code under test.
## Known limitations (stated, not papered over)
- The bash containment floor is string-matching: a cost-raiser, not a
sandbox. Quoting, encoding, `$HOME`, symlinks, or an interpreter
one-liner can evade it. The durable protection is the decision route
refusing agent-originated deciders — the filter is the belt, not the
braces.
- Approval expiry is lazy (evaluated at decide/complete/redeem), not
swept, so an expired pending row stays visible in lists until touched.
- The extension's require-approval path returns a pending message but
cannot suspend a pi session mid-turn; engine-side pause hooks cover
engine lanes only.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **Security**
* Hardened approval and permission gating with server-side decider
attribution, self-approval blocking, ownership checks, replay/race
protection, and status/TTL enforcement.
* Added fail-closed behavior for sensitive/unclassified tools and
sandbox provisioning approvals.
* Blocked credential/approval access via bash containment; plugin
destructive task operations now require explicit permission.
* **New Features**
* `fn serve` now defaults to bearer-token auth, with `--no-auth` as the
explicit opt-out.
* **Bug Fixes**
* Improved task move-source attribution (`moveSource: "user"`) and
tightened dashboard archive/bypass confirmation and operator attribution
behavior.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
104 lines
4.7 KiB
TypeScript
104 lines
4.7 KiB
TypeScript
import type { TaskStore } from "./store.js";
|
|
import type { PluginPermissions } from "./plugin-types.js";
|
|
|
|
/*
|
|
FNXC:PluginTaskStoreGate 2026-07-26-12:20:
|
|
PluginContext.taskStore historically handed every plugin the FULL TaskStore, so any
|
|
plugin could delete tasks, bypass failed pre-merge review steps, or bulk-archive the
|
|
board with no gate. This module is the smallest honest gate: a Proxy over the store
|
|
that intercepts a hardcoded denylist of destructive methods and throws unless the
|
|
plugin's manifest declares `permissions: { destructiveTaskOps: true }`. Everything
|
|
not on the denylist passes through untouched, so default plugin behavior is
|
|
otherwise unchanged.
|
|
*/
|
|
|
|
/**
|
|
* FNXC:PluginTaskStoreGate 2026-07-26-12:20:
|
|
* Destructive-method denylist. Chosen from the TaskStore surface:
|
|
* - `deleteTask` / `deleteTaskIf` / `deleteTaskById` / `deleteTaskBackend` — every
|
|
* task-deletion entry point (public and backend seams reachable via the handle).
|
|
* - `bypassFailedPreMergeReviewStep` — the FN-7720 privileged operator bypass of a
|
|
* failed pre-merge review gate; must never be callable by an ungated plugin.
|
|
* - `archiveAllDone` — the bulk archive sweep (archiveAllDone-style bulk method).
|
|
* - `cleanupArchivedTasks` — bulk destructive removal of archived task history.
|
|
* Single-task `archiveTask` is intentionally NOT denylisted: it is reversible via
|
|
* `unarchiveTask` and gating it would break benign board-hygiene plugins.
|
|
* - `getDatabase` — the raw sync SQLite handle. No in-repo plugin uses it (the QA
|
|
* plugin explicitly documents NOT to), and a raw handle would let a plugin run
|
|
* destructive SQL around the named-method denylist, so it requires the same
|
|
* destructiveTaskOps declaration.
|
|
*
|
|
* FNXC:PluginTaskStoreGate 2026-07-26-18:20:
|
|
* KNOWN RESIDUAL (review finding, deliberately not closed here): `getAsyncLayer()`
|
|
* also exposes a raw (drizzle) handle that could execute destructive SQL outside
|
|
* the denylist. It is NOT denied because four in-repo plugins (printing-press,
|
|
* compound-engineering, glasses, quality) legitimately depend on it for their own
|
|
* plugin-scoped schema/reads — denying it breaks them, and granting them
|
|
* destructiveTaskOps to compensate would defeat the gate entirely. Making this
|
|
* airtight needs a scoped/read-only data-layer design (a follow-up), not a
|
|
* denylist entry. Until then the gate is an honest guard against the named
|
|
* destructive TaskStore surface, not a sandbox for raw SQL.
|
|
*/
|
|
export const PLUGIN_DESTRUCTIVE_TASK_STORE_METHODS = [
|
|
"deleteTask",
|
|
"deleteTaskIf",
|
|
"deleteTaskById",
|
|
"deleteTaskBackend",
|
|
"bypassFailedPreMergeReviewStep",
|
|
"archiveAllDone",
|
|
"cleanupArchivedTasks",
|
|
"getDatabase",
|
|
] as const;
|
|
|
|
export type PluginDestructiveTaskStoreMethod =
|
|
(typeof PLUGIN_DESTRUCTIVE_TASK_STORE_METHODS)[number];
|
|
|
|
export interface PluginTaskStoreGateOptions {
|
|
pluginId: string;
|
|
permissions?: PluginPermissions;
|
|
}
|
|
|
|
/**
|
|
* FNXC:PluginTaskStoreGate 2026-07-26-12:20:
|
|
* Wrap a TaskStore for hand-off to a plugin context. When the plugin manifest
|
|
* declares `permissions.destructiveTaskOps: true` the raw store is returned
|
|
* unchanged. Otherwise a Proxy intercepts the denylisted methods and throws a
|
|
* clear declaration-pointing error.
|
|
*
|
|
* Implementation notes:
|
|
* - Non-denylisted function properties are bound to the RAW store (and cached per
|
|
* property) so `this` inside store methods is always the real TaskStore. This
|
|
* preserves WeakMap-keyed seams (e.g. task-move-disposer registration keyed by
|
|
* store identity) that would silently break if methods ran with the proxy as
|
|
* `this`.
|
|
* - The thrower is a plain sync function so both `store.deleteTask(...)` and
|
|
* `await store.deleteTask(...)` fail loudly.
|
|
*/
|
|
export function createPluginGatedTaskStore(
|
|
store: TaskStore,
|
|
options: PluginTaskStoreGateOptions,
|
|
): TaskStore {
|
|
if (options.permissions?.destructiveTaskOps === true) return store;
|
|
const denied = new Set<PropertyKey>(PLUGIN_DESTRUCTIVE_TASK_STORE_METHODS);
|
|
const boundMethodCache = new Map<PropertyKey, unknown>();
|
|
return new Proxy(store, {
|
|
get(target, prop) {
|
|
if (denied.has(prop)) {
|
|
return () => {
|
|
throw new Error(
|
|
`Plugin ${options.pluginId} is not permitted to call ${String(prop)}; ` +
|
|
`declare permissions.destructiveTaskOps in the plugin manifest`,
|
|
);
|
|
};
|
|
}
|
|
const value = Reflect.get(target, prop, target);
|
|
if (typeof value !== "function") return value;
|
|
const cached = boundMethodCache.get(prop);
|
|
if (cached) return cached;
|
|
const bound = (value as (...args: unknown[]) => unknown).bind(target);
|
|
boundMethodCache.set(prop, bound);
|
|
return bound;
|
|
},
|
|
}) as TaskStore;
|
|
}
|