improve-agent-instructions

improve-agent-instructions is a skill for Claude Code, Codex from Drizzy07x/Skillquiver. It costs 52 tokens per session (1,863 once invoked), scanned A, original, MIT.

A review and repair workflow for AGENTS.md and CLAUDE.md files, which contain persistent instructions for coding agents. It covers project, global, and nested instruction scopes.

In plain words
What is it for?
Use it to audit, plan, apply, or verify instruction changes for Codex and Claude Code in the permitted scopes.
Why use it?
It helps find conflicting, outdated, or unsafe agent guidance while preserving existing instructions and recoverable backups when changes are authorized.

Skill for Claude CodeCodex

Written for Claude Code and Codex: shipped in a Claude Code plugin, but also agents/openai.yaml present. Also seen: mentions CLAUDE.md; mentions Claude Code; mentions AGENTS.md.

Part of the skillquiver plugin — 24 skills shipped together

Good fit Use it to audit, plan, apply, or verify instruction changes for Codex…

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/drizzy07x/skillquiver/improve-agent-instructions
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 Drizzy07x/Skillquiver --skill improve-agent-instructions
Clone the repo
git clone --depth 1 https://github.com/Drizzy07x/Skillquiver

Made for: Claude Code, Codex.

Or install skillquiver, the plugin that ships this one along with the rest of its 24 skills.

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 improve-agent-instructions

README.md
[![agentmods](https://agentmods.dev/badge/skills/drizzy07x/skillquiver/improve-agent-instructions.svg)](https://agentmods.dev/skills/drizzy07x/skillquiver/improve-agent-instructions)
Your own site
<a href="https://agentmods.dev/skills/drizzy07x/skillquiver/improve-agent-instructions"><img src="https://agentmods.dev/badge/skills/drizzy07x/skillquiver/improve-agent-instructions.svg" alt="Measured on agentmods" height="20"></a>
Per session 52 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,863 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 $0.00052 $0.01863
Opus 5 $0.00026 $0.00932
Sonnet 5 $0.00010 $0.00373
Haiku 4.5 $0.00005 $0.00186

Measured 6d ago against content hash 169956f54214, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

improve-agent-instructions 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 6d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/inventory.mjs), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

skills/improve-agent-instructions/SKILL.md · 169 lines

How it starts

The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Improve Agent Instructions

Use an audit-first workflow. Preserve user guidance, host boundaries, and a recoverable preimage for every authorized change.

1. Resolve mode, hosts, scopes, and authorization

Modes are AUDIT, PLAN, APPLY, and VERIFY. Unknown or implicit intent resolves to AUDIT. AUDIT, PLAN, and standalone VERIFY do not write or create backups. Explicit change intent with named scopes authorizes APPLY without another confirmation.

Resolve only the named hosts and scopes (global, project, or a named nested directory) to absolute paths. Project-only and global-only requests never expand scope. Managed and resolved-external targets remain report-only. Mark a missing decision as blocked-decision; isolate it so safe independent groups can continue.

Read only requested host references before inventory:

For both hosts in one project, make shared guidance canonical. Make AGENTS.md the canonical shared project contract. Root CLAUDE.md imports @AGENTS.md; .claude/CLAUDE.md imports @../AGENTS.md. Keep only Claude-specific deltas after the import. Never copy shared rules into CLAUDE.md. Never assume that Codex expands @ imports.

For global guidance, keep Codex and Claude global files independent. Do not create cross-host global imports or treat one host's file as the other's canonical source.

2. Build the inventory

Run scripts/inventory.mjs and require schema version 1. Its stdout is the inventory contract and stderr is diagnostic. Do not silently bypass an inspector operational error. Node absence alone may use a native field-by-field fallback with unknown fields disclosed; any other inspector failure blocks the affected work.

For isolated or caller-provided roots, invoke the inspector with explicit --host, --cwd, --project, --home, --codex-home, --claude-home, and --claude-managed-dir arguments. Never substitute ambient defaults for those values.

Read the full file on GitHub · 169 lines

Files

What ships with it

4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 6d ago First seen · 169 lines · 52 tokens per session scan A 169956f54214

Subscribe to this mod's changes

improve-agent-instructions is a skill published in the GitHub repository Drizzy07x/Skillquiver (2 stars, last pushed 15d ago), licensed MIT. It adds 52 tokens to every session and 1,863 once invoked, about $0.0003 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.

Related

Other skills, from other repositories

delegate

Let Claude Architect route a versioned implementation spec through the trusted MCP runtime, independently review the Candidate Artifact, record a decision, and integrate only accepted bytes. Use for implementation delegation, Producer selection, or commitment-boundary review.

PyModel/claude-architect · 47 tokens

codex

Run the Codex CLI directly in the user's checkout for code analysis, refactoring, or automated editing without Claude Architect's verified delegation lifecycle.

PyModel/claude-architect · 31 tokens

commit

A guided process for committing code changes to Git. It checks the working tree, follows a conventional commit format, links a task when available, reviews the changes, and asks for approval.

restarter/lets-workflow · 90 tokens

orient

Internal skill for commands. Render the shared orientation snapshot - active task, what's in flight, what's next - tracker-universal. Do not trigger on user conversation; only when a command needs the snapshot.

restarter/lets-workflow · 42 tokens

actor-fetch-personality

Internal skill for commands. Fetch and validate personality from URL or file path for Actor agent. Do not trigger on user conversation - only when commands need personality loading.

restarter/lets-workflow · 37 tokens

explain

Use when user invokes /explain with a file path, directory path, function/class name, or natural language concept. Also triggers on "explain this", "how does X work", "walk me through". Produces a structured, layered explanation of what the code does, how it connects, and where to start if you need to change it. Do…

harnessprotocol/harness-kit · 95 tokens