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 agentmods add skills/dixus/claudeframework/learnnpx skills add dixus/claudeframework --skill learngit clone --depth 1 https://github.com/dixus/claudeframeworkWhat 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 | $0.00038 | $0.01838 |
| Opus 5 | $0.00019 | $0.00919 |
| Sonnet 5 | $0.00008 | $0.00368 |
| Haiku 4.5 | $0.00004 | $0.00184 |
Grade A, and why
learn 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 2d 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 — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Process new material from .claude/references/ and propose concrete changes to skills and rules. No intermediate knowledge files — findings route directly to actionable proposals.
URL ingestion (when $ARGUMENTS is a URL)
If $ARGUMENTS starts with http:// or https://, treat it as a URL to ingest:
- WebFetch the URL and extract the main content (article text, README, docs — skip nav/ads/boilerplate)
- Generate a slug from the page title or URL path (e.g.
claude-code-best-practices.md) - Write the extracted content as markdown to
references/blogs/<slug>.md - Log: "Saved → references/blogs/.md"
- Continue to the processing steps below — the new file will be picked up as unprocessed
This bridges the scout-learn pipeline: /scout produces "Worth investigating" URLs, and /learn <url> ingests them directly.
Idempotency rule — filesystem-based, read this first
Processed state is tracked on the filesystem — do NOT rely on the index table for deduplication:
- Blog files: physically moved to a
processed/subfolder - Repo directories: marked with an empty
processed/.donefile inside the repo dir (too large to move)
- Scan the filesystem for unprocessed candidate items:
references/blogs/— every file directly in this folder (NOT insideprocessed/)references/repos/— every subdirectory that does NOT have aprocessed/marker file inside it (i.e.,references/repos/<dir>/processed/.donedoes not exist)- Ignore
references/index.mdand anyprocessed/subfolders
- If there are no new items, skip to Skill review below.
Processing steps (only for unprocessed items)
For blog/article files (individual files in references/blogs/):
- Read the full content. For PDF files, use the
pagesparameter to read in chunks of 20 pages (e.g.pages: "1-20", thenpages: "21-40") — do not attempt to read an entire large PDF at once. - Extract only actionable, durable insights — skip opinions, anecdotes, and general knowledge Claude already has
- For each insight, map it to a concrete change target:
- A specific skill file (
.claude/skills/<name>/SKILL.md) — what line/section to change - A rule file (
.claude/rules/<name>.md) — create new or update existing CLAUDE.md— if it's a project-wide convention- If an insight doesn't map to any file, discard it — it's not actionable
- A specific skill file (
- Tag each insight with the affected skills:
**Affects:** [skill1, skill2] - Immediately after processing, move the file into the
processed/subfolder:references/blogs/processed/<filename>
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.
- 2d ago First seen · 109 lines · 38 tokens per session scan A 62fd59cfab49
learn is a skill published in the GitHub repository dixus/claudeframework (10 stars, last pushed 4mo ago), licensed MIT. It adds 38 tokens to every session and 1,838 once invoked, about $0.0002 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
mcporter
List, auth, and call MCP servers/tools from the terminal.
agent-messaging
Send and receive cryptographically signed messages between AI agents using the Agent Messaging Protocol (AMP). Use when the user asks to "send a message to an agent", "check agent inbox", "message another agent", "reply to a message", "notify an agent", or any inter-agent communication task.
📝 任务完成后归档
重要提醒: 每次完成复杂调试或开发任务后，主动执行此流程！ 将学到的经验归档为 skill，供以后参考。不要等用户提醒。.
oracle
Best practices for using the oracle CLI (prompt + file bundling, engines, sessions, and file attachment patterns).
agent-mode
Unified tool for managing agent LLM modes (add, remove, update, list, switch).
agento11y-prod-setup
Sets up production evaluation and guardrails for a DEPLOYED AI agent in Grafana Agent Observability, grounded in the agent's own code and its real ingested traffic. The judgment layer on top of the agento11y skill: it reads the agent's source (system prompt, tools, entrypoint) AND samples its live traffic via gcx…