host-instruction-drift-checker

host-instruction-drift-checker is a skill for Claude Code, Codex from 45ck/skill-harness. It costs 45 tokens per session (364 once invoked), scanned A, original, MIT.

A repository check that compares coding-agent instruction files across tools such as Codex, Claude, Cursor, and Copilot. It looks for differences in the rules that control agent behavior.

In plain words
What is it for?
Use it to audit agent rules, classify mismatches, find missing instruction files, and identify the correct update and regeneration path.
Why use it?
It helps prevent agents from following conflicting, outdated, or missing instructions in the same project. It also shows which file is the main source and which files should be regenerated.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: reads .claude/ paths; mentions CLAUDE.md; mentions AGENTS.md.

Good fit Use it to audit agent rules, classify mismatches, find missing instruction files, and identify the correct update and regeneration path.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/45ck/skill-harness/host-instruction-drift-checker
Install

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.

Any agent
npx skills add 45ck/skill-harness --skill host-instruction-drift-checker
Clone the repo
git clone --depth 1 https://github.com/45ck/skill-harness

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for host-instruction-drift-checker

README.md
[![agentmods](https://agentmods.dev/badge/skills/45ck/skill-harness/host-instruction-drift-checker/github.svg)](https://agentmods.dev/skills/45ck/skill-harness/host-instruction-drift-checker)
Your own site
<a href="https://agentmods.dev/skills/45ck/skill-harness/host-instruction-drift-checker"><img src="https://agentmods.dev/badge/skills/45ck/skill-harness/host-instruction-drift-checker/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.

agentmods 80×15 button for host-instruction-drift-checker

Your own site · 80×15
<a href="https://agentmods.dev/skills/45ck/skill-harness/host-instruction-drift-checker"><img src="https://agentmods.dev/badge/skills/45ck/skill-harness/host-instruction-drift-checker.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 45 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 364 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00045 $0.00364
Opus 5 $0.00023 $0.00182
Sonnet 5 $0.00009 $0.00073
Haiku 4.5 $0.00005 $0.00036

Measured 10d ago against content hash 2a8d1cb8123a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

host-instruction-drift-checker 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 10d 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.

packs/agent-operating-skills/skills/host-instruction-drift-checker/SKILL.md · 44 lines

What it actually says

Host Instruction Drift Checker

Use this skill when a repo supports multiple coding-agent hosts or when a baseline update may change agent behavior.

Process

  • Identify the canonical instruction source for the repo.
  • Inventory host surfaces:
    • AGENTS.md
    • CLAUDE.md
    • AGENT_INSTRUCTIONS.md
    • .claude/agents/*
    • .codex/agents/*
    • .agents/skills/*/SKILL.md
    • .claude/skills/*/SKILL.md
    • .codex/skills/*/SKILL.md
    • .cursor/rules/*.mdc
    • .github/copilot-instructions.md
  • Compare only the behavioral contract: scope, permissions, stop rules, evidence, generated-file ownership, and profile/installer semantics.
  • Treat host-specific syntax as normal, not drift.
  • Classify each mismatch as generated drift, overlay drift, stale doctrine, policy conflict, missing host surface, or intentional host specialization.
  • Prefer updating canonical source and regenerating derived surfaces over hand-editing generated agent files.

Output

Canonical Source

Surfaces Checked

Drift Findings

Intentional Differences

Regeneration Path

Required Manual Review

Avoid

  • forcing every host file to contain identical prose
  • editing generated files as if they were canonical
  • ignoring Cursor .mdc and Copilot surfaces when the repo advertises those hosts
  • treating missing evidence gates as minor wording drift
Changes

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.

  1. 10d ago First seen · 44 lines · 45 tokens per session scan A 2a8d1cb8123a

Subscribe to this mod's changes

host-instruction-drift-checker is a skill published in the GitHub repository 45ck/skill-harness (15 stars, last pushed 2mo ago), licensed MIT. It adds 45 tokens to every session and 364 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 2026-08-30.