Borrowing it
Nothing to install: this file belongs to PerryLink/dsh-doublecheck. 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/PerryLink/dsh-doublecheck/main/AGENTS.mdgit clone --depth 1 https://github.com/PerryLink/dsh-doublecheckWrote 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/perrylink/dsh-doublecheck/agents-md)<a href="https://agentmods.dev/instructions/perrylink/dsh-doublecheck/agents-md"><img src="https://agentmods.dev/badge/instructions/perrylink/dsh-doublecheck/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/perrylink/dsh-doublecheck/agents-md"><img src="https://agentmods.dev/badge/instructions/perrylink/dsh-doublecheck/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 | $0.02309 | $0.02309 |
| Opus 5.5 | $0.00924 | $0.00924 |
| Sonnet 5.5 | $0.00462 | $0.00462 |
| Haiku 4.5 | $0.00231 | $0.00231 |
Grade A, and why
dsh-doublecheck 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 1mo 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 — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Standalone DeepSeek Harness plugin repository (dsh-doublecheck). Development follows the dsh-plugin-guide skill and the official plugin contract; this file records repo-local decisions.
Layout
src/index.ts— package entry: the shared domain model and pure helpers. It is NOT a plugin; the installable plugin rows live at the./grilland./guardsubpath exports (seecordis.patch.yml).src/invariant.ts— the package-owned invariant companion; its standalone row loads via the./invariantsubpath export (reports write-path contradictions without the guard).src/grill/index.ts— plugin rowdoublecheck-grill: bundledgrill-requirementsskill provider + the model-facing contract tools (doublecheck_skills,doublecheck_spec,doublecheck_report) and the v0.4 verification workflow.src/guard/index.ts— plugin rowdoublecheck-guard: the policy gates (grill / tdd red-green / adversary review) on thetools/pre-executeandtools/post-executewaterfalls.src/guard/command.ts— the/doublecheck status|report|on|offsession command and the durabledoublecheck/statefold.src/guard/review.ts— adversary review orchestration (forked critic subagent, structured findings, honest "unavailable" degradation).src/guard/gate.ts— the delivery quality gate (v0.7): thegate.*Schema config, fail-loud validation, the four-phase runner (deterministic requirements/tests folds + forked consistency/local reviewers), the dsh-auto-review weak dependency (durableautoReview/*verdict records; degrade-to-local), the/gate status|run|configcommand, and the durabledoublecheck/gatefold.GateConfigSchemais the whole guard row's one live field (Config.gate, marked.volatile()insrc/guard/index.ts), so it is the row's editable settings card; the row reads the resolved block once at load, becausegate.enableddecides whether the turn-boundary notice is installed.src/guard/prose.ts— the injected reminder/deny/review/gate prose, per-language (en/zh); gate notices open with a one-sentence role statement and stay short.src/domain/— pure folds and vocabularies shared by both rows (stages, evidence, vagueness, vocabulary, report, gate). No Cordis imports.src/domain/evidence.tsownsptcSettle(), the one normalizer for settled PTC sub-dispatch events.src/events.ts— process-local Cordis event vocabulary (@mode emit, observability-only) + the durabledoublecheck/stateanddoublecheck/gateSessionEventMapmembers + this package'sMessageSourceMapmembers: the notice kinddsh-doublecheck(mixed withContextFormed) and the structureddoublecheck-review/doublecheck-gaterecords.noticeSource()andnoticeSummaryOf()live here so both guard modules build a notice, and the fold reads one back, through a single definition.skills/— four bundled discipline skills (grill-requirements,red-green-tdd,delivery-review,delivery-proof), each<name>/SKILL.mdin the generic Agent Skills layout.tests/— vitest; real CordisContextwith scripted services (subagents/commands) and synthetic durable events;tests/fixtures/holds real-transcript regression logs.scripts/— session-log tooling (decode-session,extract-fixture,scan-sessions; all take an explicit log/directory path and never read the real session store by default) + the gates and release helpers (assert-profile,check-readme-sync,verify-skills,verify-self-contained,verify-artifacts,fix-dts,release-notes,smoke-insert/verify-block/verify-report/verify-review/verify-tddpatch overlays).
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.
- 1mo ago Changed · +23 tokens per session 89b2e5305612
- 1mo ago Changed 204ec027a58b
- 1mo ago Changed · +2 lines · +430 tokens per session 3617e424a738
- 1mo ago Changed · -2 tokens per session cad661f6b0c1
- 1mo ago Changed · +1 tokens per session ce0e161c866c
- 1mo ago Changed · +1 lines · +335 tokens per session ceffab2536d2
- 2mo ago First seen · 46 lines · 1,522 tokens per session scan A 46d095844be7
dsh-doublecheck AGENTS.md is an instructions file published in the GitHub repository PerryLink/dsh-doublecheck (59 stars, last pushed 3d ago), licensed Apache-2.0. It adds 2,309 tokens to every session, about $0.0092 per session on Opus 5.5. 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.
Other instructions, from other repositories
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).
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.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
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.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.