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 MadGraphTeam/MadAgents --skill ma-reflectgit clone --depth 1 https://github.com/MadGraphTeam/MadAgentsWrote 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/madgraphteam/madagents/ma-reflect)<a href="https://agentmods.dev/skills/madgraphteam/madagents/ma-reflect"><img src="https://agentmods.dev/badge/skills/madgraphteam/madagents/ma-reflect/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/madgraphteam/madagents/ma-reflect"><img src="https://agentmods.dev/badge/skills/madgraphteam/madagents/ma-reflect.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 | $0.00171 | $0.01629 |
| Opus 5.5 | $0.00068 | $0.00652 |
| Sonnet 5.5 | $0.00034 | $0.00326 |
| Haiku 4.5 | $0.00017 | $0.00163 |
Grade A, and why
ma-reflect 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 1mo 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 — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/ma-reflect [evidence]
/ma-reflect is a self-hardening skill: point it at evidence of a past mistake and it tightens the agent system's own behaviour so the same mistake does not recur. Invoke it deliberately, with a reference to the evidence — previous runs, good and bad example pairs, a documented trap, or a recurring mistake you describe. It is not auto-invoked.
What it writes — a hardening, gated
For each failure the lead installs a hardening in whatever form fits: a sharpened operating principle, an explicit check, a routing refinement, a wiki page, a new skill, and so on. One gate governs all of them:
- Change behaviour, don't restate. A hardening that repeats a principle the surface already carries is inert — the agent already had it and the failure still happened. Before writing, confirm the surface does not already say it; if it does, the fix is to change who owns it, when it fires, or what trace it leaves, not to say it again. A hardening earns its place by being owned at the step where the failure happens and by leaving an artifact a later check can see.
- Record what worked, not only what failed. When the evidence includes an approach confirmed correct for THIS input — a good example, a path validated against source or probe — record the approved approach as a positive discipline or lesson, not only the failure it is paired against. A learned tier that holds only pitfalls pushes the system toward over-caution; the confirmations are what tell it which paths are safe to take again.
Don't harden — durable disciplines only
A real, reproduced failure is not automatically a durable discipline. Three classes are surfaced to the user and not written as permanent disciplines, because the cause is not a durable property of the work:
- Environment / instance state — a missing binary, an unset path or credential, a transient tool error. The state of this machine or run, not a rule of the task.
- A "tool/feature is broken" claim — hardening it installs a standing refusal the system will cite against itself long after the real (often environmental) cause is fixed.
- One-off specifics — a single run's exact value, name, or error string; a one-time identifier. Re-derived when needed, not memorised.
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.
- 1mo ago First seen · 59 lines · 171 tokens per session scan A bc0d528c78f3
ma-reflect is a skill published in the GitHub repository MadGraphTeam/MadAgents (14 stars, last pushed 2mo ago), licensed MIT. It adds 171 tokens to every session and 1,629 once invoked, about $0.0007 per session on Opus 5.5. 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
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…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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…