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/inth3shadows/runecho/claude-mdgit clone --depth 1 https://github.com/inth3shadows/runechoWrote 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/inth3shadows/runecho/claude-md)<a href="https://agentmods.dev/instructions/inth3shadows/runecho/claude-md"><img src="https://agentmods.dev/badge/instructions/inth3shadows/runecho/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 | $0.01023 | $0.01023 |
| Opus 5 | $0.00511 | $0.00511 |
| Sonnet 5 | $0.00205 | $0.00205 |
| Haiku 4.5 | $0.00102 | $0.00102 |
Grade A, and why
runecho CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
RunEcho Workflow
- This repo uses RunEcho as the code-truth source for symbol existence and structural drift questions.
- Use the
runechoMCP server before making claims about what functions, classes, exports, or imports exist. - If RunEcho reports stale or missing baseline data, run
runecho-ir repo reindex <name>before trusting structural answers. - Treat unresolved-symbol findings from
runecho-guardas verification stops until they are fixed or intentionally explained.
Release Tags
- Release tags must be monotonic (
vX.Y.Z, semver-increasing) — non-monotonic tags previously brokegit describe/version stamping (issue #51). - Enforced by a
pre-pushhook that rejects a tag push that isn't semver-greater than the highest existing tag. Tracked source:githooks/pre-push. Not auto-installed by git — runbash install.sh --hook-pre-pushfrom the repo root on each machine/worktree setup that pushes tags (installs into$(git rev-parse --git-common-dir)/hooks/pre-push).
Git hooks — there are five, from two different installers
Both installers write into $(git rev-parse --git-common-dir)/hooks, which is
shared across every worktree of this repo. Knowing which installer owns a hook
matters: a change that overwrites one silently disables the other's feature.
| Hook | Installed by | Does |
|---|---|---|
pre-commit |
runecho-ir install → installHooks (cmd/runecho-ir/install.go) |
runs runecho-guard at commit time |
post-commit |
same | background runecho-ir repo reindex . |
post-merge |
same | version-check --reinstall (freshness), then background runecho-ir repo reindex . |
post-checkout |
same, on branch switches only ($3 == 1) |
version-check --reinstall (freshness), then background runecho-ir repo reindex . |
pre-push |
bash install.sh --hook-pre-push (githooks/pre-push) |
rejects a non-monotonic vX.Y.Z tag push |
The three reindex hooks are the E6 auto-fresh-IR feature (#20/#21). They keep the
IR index current, and every guard answer is computed from that index — so
anything that overwrites post-merge or post-checkout degrades the guard
silently rather than loudly. The #228 freshness check (auto-reinstall when the
installed binary is behind the newest reachable tag) is folded into these SAME
two hooks — never a separate installer, which is what collided in #226 — so both
features share one hook body: freshness runs first, then the background reindex
picks up the just-rebuilt binary.
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 · 59 lines · 1,023 tokens per session scan A 4215034915e0
runecho CLAUDE.md is an instructions file published in the GitHub repository inth3shadows/runecho (0 stars, last pushed 3d ago), licensed MIT. It adds 1,023 tokens to every session, about $0.0051 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
ollama AGENTS.md
AGENTS.md instructions for ollama/ollama, covering agents.md and building.
ollama CLAUDE.md
Claude Code instructions for ollama/ollama: See AGENTS.md for the shared agent instructions for this repository.
gollem CLAUDE.md
Instructions for gollem-dev/gollem, covering claude.md, restriction & rules, commands, development and testing and test execution.
openrouter-mcp-multimodal AGENTS.md
AGENTS.md instructions for stabgan/openrouter-mcp-multimodal, covering agent instructions, before you ship, releasing (read this before publishing), short version and version files (must all match package.json).
web-researcher-mcp CLAUDE.md
Claude Code instructions for zoharbabin/web-researcher-mcp, covering claude.md — web-researcher-mcp, commands, contribution rules, architecture and non-obvious top-level dirs.
gitlab-mcp-server copilot-instructions.md
Copilot instructions for jmrplens/gitlab-mcp-server, covering gitlab-mcp-server — gitlab mcp server in go, project overview, architecture, project structure and development conventions.