Social Cue

Are scope, spend, and authority enforced while the agent runs?

Two real but thin boundaries. The agent acts through a dedicated Chrome profile separate from the operator's daily browser, and on the free tier it cannot post at all ("Nothing posts until you say so"), which is an authority limit that holds at run time. Against the rubric's top band, neither is a vendor-enforced, configurable control: the profile separation is the browser's own permission model, the posting gate is a plan boundary rather than a policy, and the paid Pro tier removes it by having Claude click submit. No spend control exists at all - the page's selling point is that work runs on the operator's existing Claude subscription with "no metered credits". Scored 1, not 0, because a boundary genuinely exists; not 2, because nothing about scope, spend or authority is enforced by the vendor.

Counts Human oversight, Operating environment, Autonomy level and Tool use, each cited below.

Is there a tamper-evident record of what it actually did?

The local dashboard keeps a weekly record of "what went out, which replies got traction". Scored 1 rather than 0 because this records what the agent actually did, which is a narrower and more relevant artefact than the user-visible output histories scored 0 on mathpal and decai (2026-10-15) - those listed the user's own creations, this lists dispatched actions. It stays at 1 because nothing is claimed about retention, export, tamper-evidence or review by anyone other than the operator, and the record lives in a local SQLite file on the operator's own machine, so there is no independent party who could ever review it.

Counts Audit trail and Deployment options, each cited below.

Are compliance obligations attached per engagement?

Nothing published. The capture contains no certification, no auditor or report, no regulatory posture, and no terms-of-service, acceptable-use or privacy document of any kind - not even a link. This is an absence found on a full read of the page, not a gap in our research. Worth a reader's attention given what the product does: it drives a signed-in browser session to post on Reddit, Hacker News and X under the operator's own identity, and submits product listings to third-party directories, none of which is addressed by any published policy.

Rests on Compliance certifications and Regulatory alignment, recorded as undisclosed and cited below. For a 0 that absence is the finding. What a score rests on

Does payment depend on a verified result?

Payment is a flat $19 per month for the Pro tier, charged whether or not a single reply lands or a single directory submission is accepted. The rubric's 0 band - paid per seat regardless of result - describes this exactly. No milestone, escrow, clawback or performance condition appears anywhere in the capture.

Rests on Pricing model and Price point, each cited below.

Can the deployment be insured, and is the customer indemnified?

No cover and no indemnity of any kind is published: no carrier, no limit, no customer indemnity, no liability cap. The absence is material rather than routine here, because the agent posts under the operator's own accounts on platforms with their own content rules, and submits listings on the operator's behalf; the operator carries the whole of that risk with nothing offered against it.

Rests on Insurance available, Insurance carriers, Coverage limits and Indemnification, recorded as undisclosed and cited below. For a 0 that absence is the finding. What a score rests on