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 bobberrisford/affiliatemcp --skill programme-anomaly-watchgit clone --depth 1 https://github.com/bobberrisford/affiliatemcpWrote 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/bobberrisford/affiliatemcp/programme-anomaly-watch)<a href="https://agentmods.dev/skills/bobberrisford/affiliatemcp/programme-anomaly-watch"><img src="https://agentmods.dev/badge/skills/bobberrisford/affiliatemcp/programme-anomaly-watch/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/bobberrisford/affiliatemcp/programme-anomaly-watch"><img src="https://agentmods.dev/badge/skills/bobberrisford/affiliatemcp/programme-anomaly-watch.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00093 | $0.01694 |
| Opus 5 | $0.00046 | $0.00847 |
| Sonnet 5 | $0.00019 | $0.00339 |
| Haiku 4.5 | $0.00009 | $0.00169 |
Grade A, and why
programme-anomaly-watch 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 12d 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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Operating instructions
You are checking the book for week-over-week anomalies. Output is a short, ranked list — nothing more.
Step 1 — enumerate the book
Call affiliate_resolve_brand with no arguments. The response is an array of { brand, network, networkBrandId } bindings.
If the array is empty, say "No brands registered — nothing to watch" and stop. Do not pad.
Step 1b - load recorded plans
Call affiliate_list_client_strategies once to see which registered brands have a plan recorded (hasStrategy / hasKpi) and whether any strategy directories are orphaned. For each registered brand in the book with either hasStrategy or hasKpi, call affiliate_get_client_strategy({ brand }) to load its strategy prose and kpi.targets. Skip registered brands with no plan; most of the book may have none, and that is fine.
This is advisory context that reshapes severity (Step 4); it never changes what the data says. Report any kpi.parseErrors verbatim and ignore those targets. If orphan strategy directories exist, report them under Failures/notes and do not invent network data for them.
Step 2 — pick the windows
Default period: the last 7 days, ending today. Comparison window: the 7 days immediately prior. Express all dates as ISO YYYY-MM-DD. Honour explicit user overrides ("this month vs last month", named dates).
Step 2b — use the snapshot for the current-window health signal
Before fanning out, you may call affiliate_build_brand_snapshot({ brand }) per brand for the current window (the 7-day default is last7d). Its value here is the count-honest byNetwork health: a network whose pull failed must not be read as "no anomalies". Treat a failed network exactly like a Step 3 binding failure — report it under Failures and say its absence of anomalies is not safe to assume.
The anomaly detection itself stays on get_programme_performance below: the week-over-week checks need the comparison window (a custom prior range the snapshot does not carry), and the top-10-dropout and silenced-publisher checks need the per-publisher rows (the snapshot's breakdown is per-programme). So the snapshot informs health, not the anomaly maths.
What ships with it
1 file 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.
- 12d ago First seen · 86 lines · 93 tokens per session scan A 860ac72c9df3
programme-anomaly-watch is a skill published in the GitHub repository bobberrisford/affiliatemcp (6 stars, last pushed 1mo ago), licensed MIT. It adds 93 tokens to every session and 1,694 once invoked, about $0.0005 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…