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/altans/herdr-cache-alert/claude-mdgit clone --depth 1 https://github.com/AltanS/herdr-cache-alertWrote 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/altans/herdr-cache-alert/claude-md)<a href="https://agentmods.dev/instructions/altans/herdr-cache-alert/claude-md"><img src="https://agentmods.dev/badge/instructions/altans/herdr-cache-alert/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 | $0.06354 | $0.06354 |
| Opus 5 | $0.03177 | $0.03177 |
| Sonnet 5 | $0.01271 | $0.01271 |
| Haiku 4.5 | $0.00635 | $0.00635 |
Grade A, and why
herdr-cache-alert 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 4d 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 — 353 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — working agreement for this repo
Cache Alert (repo AltanS/herdr-cache-alert) — a Herdr plugin that puts a
prompt-cache countdown on every agent pane and marks the turns that missed the cache. Plugin id
herdr.cache-alert (manifest: herdr-plugin.toml). Bun + TypeScript, no build step. Orientation:
README.md.
The claim contract — MANDATORY
No bare cache constant may enter src/. Every TTL, minimum-token count and cost multiplier is a
Sourced<T> (see src/claims.ts) carrying a value, a confidence, a
documentation URL, the ISO date it was checked, and a verbatim quote from that page.
When you add or change one:
- Fetch the page. Not a search result, not a memory, not another agent's summary. If you cannot open it, you cannot cite it.
- Quote verbatim. If the sentence cannot be quoted, the claim is not
documented— downgrade it toreportedorinferredand say innoteexactly what you could and could not confirm. - Stamp
retrievedAtwith the real date. This is the load-bearing field;claims --staleexists entirely to make it rot loudly. - Record gaps as gaps.
codex.*carries an explicit "OpenAI publishes no Codex-specific TTL" note. An honest hole is worth far more than a plausible number, because the plugin's whole value is that its numbers can be trusted.
confidence: "observed" is reserved for something MEASURED on this machine from the harness's own
telemetry. It outranks every documented rule. Never use it for anything you reasoned your way to.
Versioning — MANDATORY
Cache Alert is SemVered, and the version is enforced, so it never silently drifts.
The version lives in two files that must always agree, plus a matching CHANGELOG entry:
herdr-plugin.toml (canonical — Herdr reads it) · package.json · newest ## [x.y.z] heading in
CHANGELOG.md.
Before committing any functional change (anything under src/, scripts/, bin/, or the
manifest) you MUST:
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.
- 4d ago First seen · 353 lines · 6,354 tokens per session scan A 3f5645ef5dfe
herdr-cache-alert CLAUDE.md is an instructions file published in the GitHub repository AltanS/herdr-cache-alert (2 stars, last pushed 11d ago), licensed MIT. It adds 6,354 tokens to every session, about $0.0318 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
docvet CLAUDE.md
Claude Code instructions for Alberto-Codes/docvet, covering claude.md, project overview, build & development, install in development mode and with optional griffe support.
coograph brutal-honesty.instructions.md
Global communication style — applies to all workspaces and conversations.
claude-bridge AGENTS.md
Instructions for kadaba/claude-bridge, covering agents.md — using claude-bridge from any ai coding cli, two independent channels, 1. coordination (the conversation), 2. file transfer (the bytes) — same for every tool and workflows.
agent-notify AGENTS.md
AGENTS.md instructions for Akash97p/agent-notify, covering agentnotify agent workflow, start or resume work, local git workflow, required quality gates and architecture and security guardrails.
tidyfactor-design AGENTS.md
Instructions for alwkala/tidyfactor-design, covering ⚡ skill & 24 modular slash commands (7 lifecycle stages) and critical architecture (non-negotiable).
claude-code-second-brain AGENTS.md
Instructions for shofer-dev/claude-code-second-brain, covering agents.md, read first, invariants and working here.