Inbox Placement Monitor
Track where sent emails land and detect provider-level placement and sender-reputation regressions.
The skill clearly limits itself to post-send monitoring, treats exports as untrusted, labels metric provenance, emits NEEDS_INPUT for missing providers, and asks for confirmation before saving results. Resend mutation is dry-run by default and requires an explicit live flag. Deductions apply because it writes to memory, hot-cache, and open-loops without a skill-specific rollback or recovery procedure; runtime memory is not encrypted, and publisher identity is unverified.
Inputs, outputs, thresholds, baseline comparison, failure conditions, and handoff boundaries are detailed and largely self-consistent, with an explicit prohibition on inventing metrics. Deductions apply because referenced scripts, contracts, and cross-skill dependencies were not included in the supplied evidence, and there are no skill-specific tests, input-schema checks, or reproducible execution results; static calibration caps this at 10.
Trigger phrases, post-send scope, mailbox-provider coverage, required inputs, expected outputs, and non-fit boundaries are clearly stated, with English and Chinese descriptions. Deductions apply because the workflow depends on seed inbox tests and Gmail Postmaster/SNDS data, some overseas services may be unreachable from mainland China, and evidence for broader environment fit and boundary precision is limited.
The skill has frontmatter, a version, Apache-2.0 licensing, Quick Start examples, a contract, data-source notes, save rules, references, and next-skill routing, with good progressive disclosure. Deductions apply because no skill-specific changelog, maintenance owner, or update path is provided; SECURITY.md says the supported current line is 17.x while this skill declares 18.0.0, creating governance inconsistency.
The core task is concrete and the expected report is directly usable: provider placement, reputation trend, send-over-send regressions, and a SEND-S snapshot, with explicit missing-data and provenance labels. Deductions apply because static files cannot verify the calculations, formatting, or cross-source handling in practice; the workflow requires several user exports and manual inbox inspection, and contains no execution evidence. Static calibration therefore caps this at 7.
The checklist supplies explicit thresholds, fields, comparison rules, and Measured/User-provided/Estimated distinctions, while repository-level tests and CI provide some supporting evidence. Deductions apply because no dedicated tests cover this skill's key paths, no third-party execution evidence is supplied, and the referenced supporting files were not included in full; static calibration caps this at 5.
- Reconfirm scope, recipients, and recovery options before using Resend --live or any real-send capability; the skill does not specify an undo mechanism.
- Postmaster, SNDS, and seed-inbox data may contain sensitive domain, IP, complaint, and delivery information; verify memory storage and access boundaries before saving it.
- Resolve the inconsistency between SECURITY.md's supported 17.x line and the skill's 18.0.0 version, and verify that referenced files and scripts are available in the target host.
- For mainland-China use, verify access to Postmaster/SNDS dashboards and seed inboxes in advance so unavailable services are not mistaken for normal or missing telemetry.
What does this skill do, and when should you use it?
Inbox Placement Monitor is the post-send delivery skill in Aaron Marketing Skills’ email workflow. It analyzes seed-list results across Gmail, Outlook/Microsoft, Yahoo, and Apple, then trends Gmail Postmaster Tools and Microsoft SNDS reputation data against a prior send baseline. It identifies placement declines, reputation downgrades, and complaint-rate changes, and produces a reusable SEND-S placement snapshot for the email quality gate. It does not run pre-send SPF/DKIM/DMARC checks or calculate the full EQS.
Reads the sending domain, SEND profile, seed-list placement results, Gmail Postmaster Tools export, Microsoft SNDS export, and prior-send baseline; reports inbox, spam, and promotions placement by mailbox provider; reads domain/IP reputation, spam-rate, feedback-loop, IP-status, complaint-rate, and trap-hit trends; calculates send-over-send deltas and names regressions with their magnitude; evaluates only placement-related SEND-S items; labels every metric Measured, User-provided, or Estimated; and emits a placement/reputation report plus a reusable SEND-S snapshot. The skill also references an optional Resend Python helper for sending seed messages, while the actual inbox/spam/promotions verdict remains a manual mailbox check.
- A lifecycle marketer wants to verify where the latest promotional send landed at Gmail, Outlook, Yahoo, and Apple.
- A deliverability operator has Postmaster Tools and SNDS exports and needs to identify whether domain or IP reputation is improving or declining.
- An email team wants to compare the latest seed test with the previous send and quantify which provider regressed.
- A quality owner needs a post-send placement snapshot to pass into email-quality-auditor for a later SEND/EQS decision.
What are this skill's strengths and limitations?
- Clearly scoped to actual post-send placement and reputation trends.
- Requires provider-by-provider reporting for Gmail, Outlook/Microsoft, Yahoo, and Apple.
- Combines seed-list placement with Gmail Postmaster Tools and Microsoft SNDS signals.
- Separates Measured, User-provided, and Estimated metrics and does not pass missing evidence by default.
- Produces a SEND-S snapshot that can be consumed by email-quality-auditor.
- Depends on user-supplied seed tests and Postmaster/SNDS exports for a complete read.
- The actual inbox, spam, or promotions placement must still be inspected manually; the Resend helper only assists with sending and delivery-event retrieval.
- Does not perform pre-send SPF/DKIM/DMARC checks, calculate full EQS, or enforce SEND vetoes.
- The supplied material provides no independent test-suite or verified platform-coverage evidence for this individual skill.
How do you install this skill?
Use the repository on an Agent Skills-compatible host with npx skills add aaron-he-zhu/aaron-marketing-skills, or clone it with git clone https://github.com/aaron-he-zhu/aaron-marketing-skills. The skill file is email/deliver/inbox-placement-monitor/SKILL.md. The README does not document a dedicated single-skill install command for this skill.
How do you use this skill?
After a send, provide the sending domain, one SEND profile (promotional, retention, cold-outbound, or newsletter), provider-level seed-list placement results, Gmail Postmaster Tools export, Microsoft SNDS export, and a prior-send baseline. Example: Track inbox placement for example.com after my last send. Here is my seed-list test and my Gmail Postmaster + Microsoft SNDS export. Missing provider data must be reported as NEEDS_INPUT. Without a prior baseline, record the current run as the baseline rather than fabricating a delta.
How does this skill compare with similar options?
Unlike deliverability-qa, this skill handles post-send placement and reputation trends, while deliverability-qa performs the pre-send authentication and reputation pre-flight. Unlike email-quality-auditor, it emits only the SEND-S placement snapshot and does not calculate full EQS or enforce vetoes. send-experiment-designer is used for designing or reading the next send experiment.