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 instructions/dimamak/sdp/claude-mdgit clone --depth 1 https://github.com/dimamak/sdpWrote 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/instructions/dimamak/sdp/claude-md)<a href="https://agentmods.dev/instructions/dimamak/sdp/claude-md"><img src="https://agentmods.dev/badge/instructions/dimamak/sdp/claude-md.svg" alt="Measured on agentmods" 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 | $0.00220 | $0.00220 |
| Opus 5 | $0.00110 | $0.00110 |
| Sonnet 5 | $0.00044 | $0.00044 |
| Haiku 4.5 | $0.00022 | $0.00022 |
Grade A, and why
sdp CLAUDE.md 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 3d 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.
What it actually says
CLAUDE.md
Agent guides (read before matching work)
Before debugging or changing anything on the server: read docs/agent-guides/server-ops.md — restart procedure (no sudo), and rules on secrets, verification, diagnosis-only mode.
Before any deploy to /opt/dailypost/app: read
docs/agent-guides/deployment.md — it's not
a git repo, deploys are manual, server fixes must be mirrored back to git.
Code exploration
Use the codegraph MCP tool instead of Grep/Glob for symbol/code-structure
lookups — it understands the AST (definitions, call chains) while Grep only
matches text. Load it via ToolSearch (select:mcp__codegraph__codegraph_explore)
if not already available. Grep/Glob are still fine for non-symbol text (env
vars, log strings, config keys) or plain file-path search.
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.
- 3d ago First seen · 20 lines · 220 tokens per session scan A 6033dbb046e9
sdp CLAUDE.md is an instructions file published in the GitHub repository dimamak/sdp (5 stars, last pushed 5d ago), licensed MIT. It adds 220 tokens to every session, about $0.0011 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 instructions, from other repositories
last30days-skill AGENTS.md
AGENTS.md instructions for mvanhorn/last30days-skill, covering last30days skill, structure, orientation, commands and prefer github actions → "prepare release". local equivalent.
last30days-skill copilot-instructions.md
Copilot instructions for mvanhorn/last30days-skill, covering copilot-specific guidance, test generation, pull request reminders, vendor exclusion zone and ci expectations.
last30days-skill CLAUDE.md
Claude Code instructions for mvanhorn/last30days-skill, a project described as: AI agent skill that researches any topic across Reddit, X, YouTube, HN, Polymarket, and the web - then synthesizes a grounded summary.
adhx CLAUDE.md
Instructions for itsmemeworks/adhx, covering claude.md, local configuration, example claude.local.md, github cli and sentry cli.
web-search-mcp CLAUDE.md
Instructions for sydasif/web-search-mcp, covering claude.md, common commands, development, testing and architecture.
tweetclaw AGENTS.md
Instructions for Xquik-dev/tweetclaw: Each commit must reduce net handwritten source LOC. No exceptions. No one may waive this rule. Additions require larger, useful simplifications elsewhere in this repository. Formatting, minification, generated files, and deleted valuable tests or docs do not count.