Borrowing it
Nothing to install: this file belongs to sakurameishi/codex-dsh-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/sakurameishi/codex-dsh-mcp/main/AGENTS.mdgit clone --depth 1 https://github.com/sakurameishi/codex-dsh-mcpWrote 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/sakurameishi/codex-dsh-mcp/agents-md)<a href="https://agentmods.dev/instructions/sakurameishi/codex-dsh-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/sakurameishi/codex-dsh-mcp/agents-md/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.
<a href="https://agentmods.dev/instructions/sakurameishi/codex-dsh-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/sakurameishi/codex-dsh-mcp/agents-md.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.01188 | $0.01188 |
| Opus 5 | $0.00594 | $0.00594 |
| Sonnet 5 | $0.00238 | $0.00238 |
| Haiku 4.5 | $0.00119 | $0.00119 |
Grade A, and why
codex-dsh-mcp AGENTS.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 9d 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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
DeepSeek Delegation Policy
Goal
Minimize Codex usage without materially reducing correctness. DeepSeek Harness is the low-cost execution worker; Codex remains the router, architect, reviewer, and final verifier.
Route every development task
Choose DIRECT_CODEX or DELEGATE_DEEPSEEK before broad repository exploration. Do not read the whole project merely to decide which route applies.
Choose DELEGATE_DEEPSEEK for context-heavy or routine work: repository-wide search, reading more than about three files, tracing functions/call chains/dependencies, ordinary features and CRUD, routine UI edits, changes across more than about two files, mechanical edits, refactors, tests, ordinary bugs, compiler/type/lint failures, repeated edit-test-fix loops, documentation tied to code, duplicate-code cleanup, and project-structure exploration.
When context is missing, start with deepseek_explore. Use deepseek_task for ordinary implementation, deepseek_fix when concrete error/test evidence already exists, and deepseek_review for a useful independent second pass. Workers change only isolated Git worktrees; never apply a worker patch automatically.
Choose DIRECT_CODEX for tiny or obvious edits, one-line or small configuration changes, work the user explicitly asks Codex to perform personally, core architecture or public API decisions, core database-model design, security-sensitive authentication/authorization/payment logic, destructive or irreversible operations, ambiguous product decisions, final diff review, final verification, and every task already marked ESCALATED_TO_CODEX.
Retry and escalation
The bridge defaults to at most one retry (DSH_AUTO_RETRY=true, DSH_MAX_RETRIES=1): the initial attempt plus one focused bug-fix attempt for eligible non-timeout failures, with the same task_id and incremented attempt. DSH_TIMEOUT_MS and timeout_minutes are total wall-clock budgets for the whole delegation, not fresh budgets for each attempt.
Treat non-zero exits, timeouts, Harness/controller failures, malformed output, explicit inability to complete, empty expected diffs, and failed key tests as failures. A timeout is never retried automatically: it consumes the total budget and escalates immediately. Other eligible failures may receive one focused retry; after that failure the bridge returns status: "escalated_to_codex" and ESCALATED_TO_CODEX.
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.
- 9d ago First seen · 65 lines · 1,188 tokens per session scan A c5affb93d939
codex-dsh-mcp AGENTS.md is an instructions file published in the GitHub repository sakurameishi/codex-dsh-mcp (0 stars, last pushed 10d ago), licensed MIT. It adds 1,188 tokens to every session, about $0.0059 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
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.
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.
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).
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).
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.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.