dhpk-agent-architecture-audit

dhpk-agent-architecture-audit is a skill for Claude Code from hmj1026/dhpk. It costs 122 tokens per session (1,974 once invoked), scanned A, original, MIT.

A diagnostic workflow for applications that use AI agents or language models. It examines problems caused by wrapper layers, memory, tools, retries, transport, and response rendering.

In plain words
What is it for?
Use it when an agent becomes unreliable after changes to prompts, memory, tools, or multi-step workflows, or when tool calls and responses behave inconsistently.
Why use it?
It helps locate failures that are hidden by the surrounding agent system when the underlying model works correctly on its own.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

Part of the dhpk plugin — 65 skills, 31 commands, 37 agents, 3 hooks shipped together

Good fit Use it when an agent becomes unreliable after changes to prompts, memory, tools, or multi-step workflows, or when tool calls and responses behave inconsistently.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/hmj1026/dhpk/dhpk-agent-architecture-audit
View source ↗ hmj1026/dhpk
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 hmj1026/dhpk --skill dhpk-agent-architecture-audit
Clone the repo
git clone --depth 1 https://github.com/hmj1026/dhpk

Made for: Claude Code.

Or install dhpk, the plugin that ships this one along with the rest of its 65 skills, 31 commands, 37 agents, 3 hooks.

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 dhpk-agent-architecture-audit

README.md
[![agentmods](https://agentmods.dev/badge/skills/hmj1026/dhpk/dhpk-agent-architecture-audit/github.svg)](https://agentmods.dev/skills/hmj1026/dhpk/dhpk-agent-architecture-audit)
Your own site
<a href="https://agentmods.dev/skills/hmj1026/dhpk/dhpk-agent-architecture-audit"><img src="https://agentmods.dev/badge/skills/hmj1026/dhpk/dhpk-agent-architecture-audit/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.

agentmods 80×15 button for dhpk-agent-architecture-audit

Your own site · 80×15
<a href="https://agentmods.dev/skills/hmj1026/dhpk/dhpk-agent-architecture-audit"><img src="https://agentmods.dev/badge/skills/hmj1026/dhpk/dhpk-agent-architecture-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 122 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,974 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 $0.00122 $0.01974
Opus 5 $0.00061 $0.00987
Sonnet 5 $0.00024 $0.00395
Haiku 4.5 $0.00012 $0.00197

Measured 7d ago against content hash db9aec4747df, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

dhpk-agent-architecture-audit 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.

generated/claude-profiles/compat-v1/package/skills/dhpk-agent-architecture-audit/SKILL.md · 209 lines

How it starts

The opening of the file, as written. The whole thing — 209 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Architecture Audit

Source: oh-my-agent-check.

A diagnostic workflow for agent systems that hide failures behind wrapper layers, stale memory, retry loops, or transport/rendering mutations.

When to Activate

MANDATORY for:

  • Releasing any agent or LLM-powered application to production
  • Shipping features with tool calling, memory, or multi-step workflows
  • Agent behavior degrades after adding wrapper layers
  • User reports "the agent is getting worse" or "tools are flaky"
  • Same model works in playground but breaks inside your wrapper
  • Debugging agent behavior for more than 15 minutes without finding root cause

Especially critical when:

  • You've added new prompt layers, tool definitions, or memory systems
  • Different agents in your system behave inconsistently
  • The model was fine yesterday but is hallucinating today
  • You suspect hidden repair/retry loops silently mutating responses

When NOT to Use

  • General code debugging — use agent-introspection-debugging
  • Code review — use language-specific reviewer agents
  • Security scanning — use change-verdict security mode
  • Agent performance benchmarking — use agent-eval
  • Writing new features — use the appropriate workflow skill

Audit Model

Load references/layers-and-failure-patterns.md before mapping findings to the 12 layers or the common failure patterns. It is intentionally disclosed because most audits need the workflow first and the full diagnostic vocabulary only during failure mapping.

Audit Workflow

Phase 1: Scope

Define what you're auditing:

  • Target system — what agent application?
  • Entrypoints — how do users interact with it?
  • Model stack — which LLM(s) and providers?
  • Symptoms — what does the user report?
  • Time window — when did it start?
  • Layers to audit — which of the 12 layers apply?

Phase 2: Evidence Collection

Gather evidence from the codebase:

  • Source code — agent loop, tool router, memory admission, prompt assembly
  • Logs — historical session traces, tool call records
  • Config — prompt templates, tool schemas, provider settings
  • Memory files — SOPs, knowledge bases, session archives

Read the full file on GitHub · 209 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 · 209 lines · 122 tokens per session scan A db9aec4747df

Subscribe to this mod's changes

dhpk-agent-architecture-audit is a skill published in the GitHub repository hmj1026/dhpk (2 stars, last pushed yesterday), licensed MIT. It adds 122 tokens to every session and 1,974 once invoked, about $0.0006 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 2026-09-05.

Related

Other skills, from other repositories

plumb-line-remediate

Use when applying findings from a plumb-line audit report — the builder has a report (or pasted findings) and wants the fixes made. Opt-in and separate from the audit, which is read-only and never fixes.

slopstopper/plumb-line · 51 tokens

generator-apply-fixes

Internal Auto-Harness generator skill for QA fix cycles. Use only inside the Generator subagent when it is addressing named defects from QA or retest.

redker56/auto-harness · 36 tokens

debug

Interactive debugging workflow with hypothesis-driven probe loop. Use when: unknown bugs, script errors, silent failures, troubleshooting. Not for: known bugs (use bug-fix), GitHub issue analysis (use issue-analyze), code understanding (use code-explore). Output: debug report with probe journal + root cause + fix.

sd0xdev/sd0x-harness · 67 tokens

anti-patterns

Catalogue of known SDLC anti-patterns that greatcto agents must actively reject when reviewing architecture, plans, code, or post-mortems. Used by architect (pre-impl), pm (planning), senior-dev (impl), l3-support (post-incident).

avelikiy/great_cto · 59 tokens

windows-compat

Audit and harden this Rust repo (code-graph-mcp) for Windows correctness: path-spelling drift between producers, the 32,767-char command-line cap, index-key mismatches, and path predicates that assume one ecosystem's layout. Use whenever touching code that builds, compares, prints, or stores a filesystem path; that…

sdsrss/code-graph-mcp · 165 tokens

incremental-analysis

Detect existing workspace, diff current sources against high water mark metadata, classify specs as unchanged/stale/orphaned/new, re-analyze only what changed. Activates automatically when /analyze finds an existing workspace.

prime-radiant-inc/greenfield · 47 tokens