Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add swan-gtm/gtm-skills --skill weekly-performance-advisorgit clone --depth 1 https://github.com/swan-gtm/gtm-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/swan-gtm/gtm-skills/weekly-performance-advisor)<a href="https://agentmods.dev/skills/swan-gtm/gtm-skills/weekly-performance-advisor"><img src="https://agentmods.dev/badge/skills/swan-gtm/gtm-skills/weekly-performance-advisor/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/swan-gtm/gtm-skills/weekly-performance-advisor"><img src="https://agentmods.dev/badge/skills/swan-gtm/gtm-skills/weekly-performance-advisor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00131 | $0.01448 |
| Opus 5 | $0.00066 | $0.00724 |
| Sonnet 5 | $0.00026 | $0.00290 |
| Haiku 4.5 | $0.00013 | $0.00145 |
Grade A, and why
weekly-performance-advisor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Applies to a book of running campaigns on a recurring cadence. Produces a reply queue in urgency order, a ranked fix list, and the week's movement.
Two questions, in this order
A weekly review answers who is waiting on me and what is broken, and the first outranks the second every time. A prospect who replied four days ago and got nothing is a lost deal that already happened; a campaign three points below benchmark is a problem that will still be there tomorrow.
Most dashboards invert this, leading with aggregate performance because it's easier to compute. Lead with the people.
Nothing is scored below ten sends
A campaign with six sends and one reply does not have a 17% reply rate. It has six sends.
Set a floor — ten sends per metric before it gets a status at all — and render everything under it as insufficient volume rather than as a number. Without that floor, the smallest and newest campaigns dominate every ranking, and the weekly review turns into a tour of statistical noise.
The same rule applies per step, not just per campaign: a sequence can have plenty of total volume and a fourth touch that only twelve people ever reached.
Aggregate on the ratio, not on the percentages
Portfolio-level numbers are where this goes quietly wrong. Averaging campaign reply rates gives a 2,000-send campaign and a 40-send campaign equal weight in the number leadership reads.
Sum the numerators, sum the denominators, divide once. A volume-weighted portfolio rate is the rate that actually happened; a mean of percentages is an artefact of how the work was split into campaigns.
Judge the campaign on its primary channel
A multichannel campaign that sends 900 LinkedIn messages and 60 emails is a LinkedIn campaign. Letting the email leg's weak numbers flip the whole campaign to "broken" sends someone off to rewrite copy that 6% of the audience saw.
Identify the primary channel by volume, treat a channel as co-primary only when it carries a meaningful share, and let only the driving channel's metrics set the campaign's status. A weak secondary channel is worth flagging as a leg to fix — not worth condemning the campaign for. Technical health is the exception: a bounce problem counts regardless of channel weighting, because it damages the sending domain rather than just this campaign.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 84 lines · 131 tokens per session scan A 44c08515f538
weekly-performance-advisor is a skill published in the GitHub repository swan-gtm/gtm-skills (150 stars, last pushed 2d ago), licensed MIT. It adds 131 tokens to every session and 1,448 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
huashu-bento-insight
OpenDesign vs closed cloud design tools: a side-by-side displacement case on control, cost (BYOK), and lock-in. Built as a decision-grade B2B sales deck for evaluation committee.
html-ppt-product-launch
OpenDesign Teams: a launch-and-adoption proposal for a mid-market design team weighing a switch from closed cloud tools. Built as a decision-grade B2B sales deck for design team lead, IT.
afrexai-lead-hunter
Enterprise-grade B2B lead generation, enrichment, scoring, and outreach sequencing for AI agents. Find ideal prospects, enrich with verified data, score against your ICP, and generate personalized outreach — all autonomously.
deliverability-monitoring
Monitor outbound deliverability health across domains, inboxes, DNS, bounces, spam placement, reply rates, and reputation signals. Produces alert thresholds, dashboard spec, weekly QA checklist, and remediation playbook. Use when inboxing drops, bounce rates rise, campaigns underperform, or new infrastructure launches.
icp-scoring
Score and tier ICP-fit accounts using firmographic, technographic, intent, pain, and timing signals. Produces a weighted ICP scorecard, tiering rules, anti-ICP exclusions, routing thresholds, and validation checklist. Use when defining target accounts, ranking TAM lists, prioritizing ABM accounts, or building fit…
pricing-strategy
Design SaaS pricing strategy using willingness-to-pay research, packaging, value metrics, competitive anchors, and expansion paths. Produces pricing hypotheses, package architecture, discount guardrails, and test plan. Use when setting prices, changing packages, evaluating usage-based pricing, or fixing monetization.