aidoc-flow-framework: Instructions file for Claude Code

CLAUDE.md

aidoc-flow-framework CLAUDE.md is an instructions file for Claude Code from vladm3105/aidoc-flow-framework. It costs 17,564 tokens per session, scanned A, original, MIT.

A set of Claude Code project instructions stored in a file named CLAUDE.md. It records the project's purpose, shared rules, development workflow, and relationship between its platforms.

In plain words
What is it for?
Use it to orient agents working on the AI document-flow framework, its Hermes MCP server, its Claude Code plugin, and their shared conformance tests. It also describes the project's specification flow.
Why use it?
It gives coding agents durable project context so they can follow the same conventions across sessions and keep the framework and its implementations aligned.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions CLAUDE.md; names the TodoWrite tool; mentions Claude Code.

This is vladm3105/aidoc-flow-framework's own configuration. It tells Claude Code how to work on aidoc-flow-framework 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 aidoc-flow-framework configures →

Reuse

Borrowing it

Nothing to install: this file belongs to vladm3105/aidoc-flow-framework. 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/vladm3105/aidoc-flow-framework/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/vladm3105/aidoc-flow-framework

Made for: Claude Code.

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 aidoc-flow-framework CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/vladm3105/aidoc-flow-framework/claude-md.svg)](https://agentmods.dev/instructions/vladm3105/aidoc-flow-framework/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/vladm3105/aidoc-flow-framework/claude-md"><img src="https://agentmods.dev/badge/instructions/vladm3105/aidoc-flow-framework/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 17,564 This file is loaded in full into every session.
When invoked 17,564 The same file — it is already loaded in full.
Security scan A 1 finding. Scan, not verified.
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 $0.17564 $0.17564
Opus 5 $0.08782 $0.08782
Sonnet 5 $0.03513 $0.03513
Haiku 4.5 $0.01756 $0.01756

Measured yesterday against content hash 209ca6ad7149, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

aidoc-flow-framework CLAUDE.md scanned grade A with 1 finding 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 yesterday.

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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

not a tooling requirement (both `curl` a pinned static binary). This means the
CLAUDE.md · 986 lines

How it starts

The opening of the file, as written. The whole thing — 986 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md — Project Memory

Persistent context for the AI Doc Flow Framework. Auto-loaded every session. Keep it short and current.

Non-Claude agents (Codex, Gemini CLI, Copilot, Hermes) start at AGENTS.md — the short cross-agent orientation. This file remains the full working agreement; where the two disagree, this one wins.

What this project is

The document-flow framework, delivered as one engine-agnostic specification (framework/) with two independent platforms:

  • Platform A — Hermes AI — MCP-server engine (platforms/hermes/).
  • Platform B — Claude Code plugin — native Claude Code engine, no MCP (platforms/claude-code-plugin/).

The platforms share the framework/ spec and nothing else. Both pass the same shared conformance suite (tests/conformance/). The framework/ spec defines the 8-layer SDD flow (BRD → PRD → EARS → BDD → ADR → SPEC → TDD → IPLAN → Code).

Current state (as of 2026-08-16): framework spec 0.51.0, Claude Code plugin 0.25.0 (pre-1.0 preview, 52 skills = 50 active + 2 deprecated stubs), Hermes 0.12.1. YAML-BDD arc complete (BDD authored as structured scenarios: YAML, not Gherkin-in-markdown) and the CONSUMER-FEEDBACK P1 wave shipped: element-level COV01/COV02 coverage (D-0039 — REALIZING_LAYERS map; catches orphaned requirement elements), manual-mode provisional IDs + normative SHA-256 algorithm (D-0040 — id_state/PROV01; element IDs are LLM-generated stable strings, NOT verified content-hashes), and first-class reuse / satisfied-by-reference (D-0041 — reuse: frontmatter; REUSE01/REUSE02). PROVISIONAL-IDS-002 Phase 1 shipped (D-0061/D-0062, spec 0.35.0): the element-ID hash-input contract (normalization transform + BRD §7 extraction boundary) is formalized in ID_NAMING_STANDARDS.md, and python -m sdd_doc_lint.rehash --check verifies a canonical BRD's §7 FR IDs against it on demand (IDDRIFT01 — advisory, opt-in, NOT in the default lint). Scoped "verifiable on demand," not "verified"; rehash --fix + all-layer extraction + corpus reconciliation are founder-decided Phase 2+. ELEMENT-ID-LAYER-CONTRACT-001 shipped (GD-09/D-0067, spec 0.39.0): that transform had reached only BRD-TEMPLATE.yaml, so the re-specified algorithm is now deleted from the other four layer templates + three layer READMEs in favour of a cross-reference to ID_NAMING_STANDARDS.md; TDD gains the element-ID contract it never had; the inert placeholder: "0000" key is removed; tests/conformance/test_element_id_layer_contract.py locks all of it over framework/layers/**spec only at the time, leaving the 19 plugin/Hermes authoring surfaces (#342) and the acceptance harness's second hash implementation (#351) open. Both have since closed (2026-07-26/27): the element-ID generator shipped as python -m sdd_doc_lint.rehash --compute (PR #363, D-0068), and the acceptance harness now delegates to that one implementation (PR #366). The guard that locks the negative property (tests/conformance/platforms/test_no_inprompt_hashing.py) scans less than its name implies and one unscanned surface still hashes — #385 closed the plugin-SKILL and Hermes-reference halves, but agent-skills/**/SKILL.md is still reached by no root — so a green run of it is not evidence that no surface hashes. Plugin also ships full 8-layer playbook injection + preemptive saga driver across all 8 autopilots (SAGA-PARITY-001) + per-layer model-recommendation precheck (MODEL-PRECHECK-ROLLOUT) + review-quality calibration + necessary-upstream contract (NECESSARY-UPSTREAM-001) + threshold-resolution gate (TH-RES-001) + per-PR doc-of-record discipline (DOC_GOVERNANCE_CORE.md Principle 8). 8-layer development sequence complete. Hermes has since advanced substantially (from 0.7.3): the audit_threshold raise-only gate (HERMES-ADAPT-ENFORCE-001), .aidoc/profile.yaml runtime consumption, and the opt-in bounded review→remediate→re-review quality loop (HERMES-REVIEW-LOOP-001 Phase 1, D-0063). Residual arc: Hermes parity — remaining plugin-vs-Hermes deltas + quality-loop Phase 2 (cross-invocation resume / G-R1, parallel-review lock fix), tracked in plans/HERMES-BACKLOG.md. The example corpus is regenerated wholesale after framework changes (so corpus-remediation findings are deferred to that regen). IPLAN ↔ iplanic integration deferred — see plans/IPLAN-IPLANIC-DEFERRED.md.

Read the full file on GitHub · 986 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. yesterday Changed · +20 lines · +487 tokens per session 209ca6ad7149
  2. 2d ago Changed · +59 lines · +1,233 tokens per session 646ae8f28b52
  3. 5d ago First seen · 907 lines · 15,844 tokens per session scan A 29a9de045f16

Subscribe to this mod's changes

aidoc-flow-framework CLAUDE.md is an instructions file published in the GitHub repository vladm3105/aidoc-flow-framework (17 stars, last pushed yesterday), licensed MIT. It adds 17,564 tokens to every session, about $0.0878 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

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.

vercel/next.js · 7,296 tokens

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 · 5,182 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

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

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

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