agentmods.dev

agent-export: Instructions file for Claude Code

CLAUDE.md

agent-export CLAUDE.md is an instructions file for Claude Code, Codex from xiaojiou176-open/agent-export. It costs 2,138 tokens per session, scanned A, original, MIT.

A set of Claude Code instructions for the agent-export repository. It documents the repository's current design, important files, supported exports, search tools, reports, and integration checks.

In plain words
What is it for?
Use it when working on agent-export, especially its Rust command-line exporter, archive and report features, retrieval tools, integration diagnostics, and governance material.
Why use it?
It gives the coding agent a shared reference for how this repository works and where its documentation and commands live.

Instructions file for Claude CodeCodex

Written for Codex and Claude Code: reads ~/.codex or $CODEX_HOME, but also the file is CLAUDE.md. Also seen: reads .claude/ paths; mentions Claude Code; mentions AGENTS.md.

This is xiaojiou176-open/agent-export's own configuration. It tells Claude Code and Codex how to work on agent-export itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything agent-export configures →

not rated 1repo 1mo ago A scan 2,138 tokens original MIT
View source ↗ xiaojiou176-open/agent-export xiaojiou176-open.github.io
Reuse

Borrowing it

Nothing to install: this file belongs to xiaojiou176-open/agent-export. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/xiaojiou176-open/agent-export/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/xiaojiou176-open/agent-export

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 agent-export CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/xiaojiou176-open/agent-export/claude-md.svg)](https://agentmods.dev/instructions/xiaojiou176-open/agent-export/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/xiaojiou176-open/agent-export/claude-md"><img src="https://agentmods.dev/badge/instructions/xiaojiou176-open/agent-export/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 2,138 This file is loaded in full into every session.
When invoked 2,138 The same file — it is already loaded in full.
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 $0.02138 $0.02138
Opus 5 $0.01069 $0.01069
Sonnet 5 $0.00428 $0.00428
Haiku 4.5 $0.00214 $0.00214

Measured 7d ago against content hash 196a43f306b5, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

agent-export 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 7d 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.

CLAUDE.md · 143 lines

How it starts

The opening of the file, as written. The whole thing — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AgentExport CLAUDE Guide

Quick Index

  1. AGENTS.md
  2. README.md
  3. docs/README.md
  4. docs/adr/ADR-0001-source-layering.md
  5. docs/adr/ADR-0002-codex-first-delivery.md
  6. docs/reference/host-safety-contract.md

Current Truth

  • This repo is a Rust CLI-first exporter.
  • Current implementation delivery is Codex dual-source export + minimal Claude Code session-path export + shared JSON/HTML export + local archive index + local metadata search + semantic retrieval + hybrid retrieval + local multi-agent archive shell + local retrieval report artifacts + workspace-local transcript backlinks + local reports shell + reports-shell metadata search + repo-owned integration materializer/doctor + integration doctor hardening + platform-aware integration doctor diagnostics + integration pack-shape hardening + integration onboarding experience + integration evidence pack + integration evidence shell search + machine-readable integration evidence + integration evidence timeline/diff + evidence gate / explain + baseline registry + policy packs + promotion engine/history + remediation bundle studio + read-only governance MCP surface + current-decision automation + local governance workbench.
  • Current integration-pack semantics:
    • Codex and Claude Code are ready through CLI-first templates plus an optional minimal stdio MCP bridge
    • the MCP bridge resolves repo-local launcher paths before any explicit AGENT_EXPORT_BIN / AGENT_EXPORT_ARGS override
    • OpenClaw is prepared as bundle content and plugin skeletons, not as a repo-native runtime
    • integrate <platform> --target <dir> materializes repo-owned assets into explicit targets only
    • doctor integrations --platform <platform> --target <dir> stays read-only and reports ready / partial / missing
    • doctor now also checks target-content drift against the current repo-generated content and runs a launcher probe
    • if the launcher can only fall back to cargo run, doctor keeps the result conservative instead of triggering a build in read-only mode
    • doctor now also validates platform-specific config/bundle shape for Codex, Claude Code, and OpenClaw
    • doctor now also tightens Codex command/args shape and Claude pack-shape diagnostics
    • onboard <platform> --target <dir> now stitches materialize + doctor + next-step guidance into one first-run path
    • integrate / onboard now reject obvious live host/global roots such as ~/.codex, ~/.claude*, and direct OpenClaw bundle/plugin roots
    • doctor integrations --save-report and onboard --save-report now write static integration evidence under .agents/Integration/Reports
    • integration evidence reports are integration-owned artifacts and must not be re-ingested into transcript/search corpora
    • the integration evidence front door now supports static shell search/facets for platform and readiness
    • integration evidence now also writes paired report.html + report.json artifacts plus index.html + index.json
    • AgentExport evidence diff --left <report> --right <report> now compares saved integration evidence snapshots without rerunning doctor/onboard
    • AgentExport evidence gate --baseline <report> --candidate <report> now classifies evidence as pass / warn / fail
    • AgentExport evidence explain --report <report> and doctor integrations --explain now expose remediation order
    • AgentExport evidence baseline list|show|promote now manages the official baseline registry
    • AgentExport evidence policy list|show now exposes repo-owned governance policy packs
    • AgentExport evidence promote / evidence history now record decision promotion and auditable governance history
    • the MCP bridge now also exposes read-only governance consumption tools
  • The repository is designed to grow into multiple connectors later, but not all at once.
  • Current export semantics stay aligned with CodexMonitor:
    • thread/read primary
    • thread/resume fallback only
    • fallback exports are incomplete
    • Markdown stays round-based and remains the default output
    • output target semantics remain explicit
  • Current local direct-read semantics:
    • --source local
    • thread-id -> sqlite -> rollout_path or direct --rollout-path
    • result is degraded
    • local source is archival truth, not canonical truth
  • Current Claude Code semantics:
    • export claude-code --session-path <PATH>
    • local session-file import into the shared archive core
    • result is degraded
    • Claude Phase 3 is a second-connector proof, not a second Markdown dialect
  • Current output-format semantics:
    • --format markdown|json|html
    • markdown stays the default
    • json is a second output over the same transcript core
    • html is a third output over the same transcript core
    • html is a static readable transcript document, not a browse shell
    • workspace conversations HTML now includes a backlink to the local archive shell
    • Downloads HTML stays free of workspace-only links
  • Current archive browsing / publish semantics:
    • publish archive-index --workspace-root <repo>
    • scans workspace conversations for existing HTML transcript exports
    • writes one local archive shell index.html with relative links
    • now supports local metadata filtering plus retrieval-lane guidance for semantic / hybrid CLI search
    • now also writes a local reports shell at .agents/Search/Reports/index.html
    • local reports shell now also supports report search and report-kind filtering
    • keeps semantic / hybrid retrieval in CLI instead of moving them into browser-side execution
    • does not add hosted publish / remote search behavior
  • Current semantic retrieval semantics:
    • search semantic --workspace-root <repo> --query "<text>"
    • uses embedding-based retrieval over the local archive corpus
    • requires local model assets for live retrieval
    • does not silently fall back to lexical search
  • Current hybrid retrieval semantics:
    • search hybrid --workspace-root <repo> --query "<text>"
    • blends semantic ranking with lexical metadata signals
    • reuses the persistent semantic index from the semantic path
    • does not mutate search semantic into a hidden hybrid command
  • Current retrieval report semantics:
    • search semantic --save-report and search hybrid --save-report
    • write static HTML reports under .agents/Search/Reports
    • reports are search-owned local artifacts, not transcript HTML inputs
    • archive shell and reports shell may link them, but retrieval execution stays in CLI
  • Current highest-value next step:
    • a new post-Phase-34 product decision, still local-first and non-hosted
  • Current host-safety semantics:
    • the repo may spawn one direct app-server child
    • the repo may only terminate that directly owned child handle
    • host-control utilities, shell launchers, desktop automation, and inline-eval launcher overrides are rejected

Read the full file on GitHub · 143 lines

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. 7d ago First seen · 143 lines · 2,138 tokens per session scan A 196a43f306b5

Subscribe to this mod's changes

agent-export CLAUDE.md is an instructions file published in the GitHub repository xiaojiou176-open/agent-export (1 stars, last pushed 1mo ago), licensed MIT. It adds 2,138 tokens to every session, about $0.0107 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.

Claude Codeai-agentsarchive-workbenchclicodexevidence-routing
Related

Other instructions, from other repositories

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,469 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 tokens