Agent Architecture Audit
Diagnose a misbehaving agent across 12 stack layers, from system prompt to hidden repair loops, with severity-ranked findings.
- What
- Diagnose a misbehaving agent across 12 stack layers, from system prompt to hidden repair loops, with severity-ranked findings.
- Cost
- Free
- Needs
- the Muse app (mobile or web), and access to the agent system you want to audit (codebase, logs, configs). No installation, no API key.
- Level
- Beginner — the Muse app (mobile or web)…
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
Curated by Skill Harbor: a full-stack diagnostic workflow for agent and LLM applications that hide failures behind wrapper layers. It audits 12 layers (system prompt, session history, long-term memory, distillation, active recall, tool selection, tool execution, tool interpretation, answer shaping, platform rendering, hidden repair loops, persistence) against five failure patterns: wrapper regression (the model is fine, the wrapper makes it worse), memory contamination (old topics leaking into new conversations), tool discipline failure ("must use tool X" only in prompt text, never enforced in code), rendering or transport corruption, and hidden agent layers silently mutating output. Four-phase workflow: scope the audit, collect evidence (with rg recipes for each anti-pattern), map failures (symptom, mechanism, source layer, root cause, evidence, confidence), then a code-first fix strategy ordered by severity (critical, high, medium, low) with a structured JSON report schema. Includes 7 quick diagnostic questions and a blunt reporting rule: if the system is broken, say so directly. It references sibling ECC skills (agent-introspection-debugging, agent-eval, security-review, autonomous-agent-harness, agent-harness-construction), which are separate skills not all listed here. By @affaan-m, listed here with credit to its creator. From the affaan-m/ECC repository (MIT). Honest caveats: built for developers with codebase access; inside Muse it works as a diagnostic interview and audit framework rather than automated scanning. Skill Harbor never reviews the code, review it yourself before use.
Version:
Quick answer
Agent Architecture Audit is a free Muse skill that diagnoses a misbehaving agent across twelve stack layers, from system prompt to hidden repair loops, with severity-ranked findings. It audits system prompt, session history, long-term memory, recall, tool selection, execution and interpretation, answer shaping, rendering and persistence, against failure patterns like wrapper regression, memory contamination and tool discipline failure.
Install
Copy the install package below, then paste it into Muse🧪 Install ran in our test sandbox · Nov 11, 2026📌 Pinned version: commit 4eb71d9Pinned version: install the code at commit 4eb71d92a39cab44ad40ac9d8a6a5ccb4029d6c2 exactly (Skill Harbor tested this version). Do not substitute a newer version. If this commit is unreachable, stop and tell me. Use "Agent Architecture Audit" with your Muse. Prerequisites: the Muse app (mobile or web), and access to the agent system you want to audit (codebase, logs, configs). No installation, no API key. 1. Open the skill: https://raw.githubusercontent.com/affaan-m/ECC/4eb71d92a39cab44ad40ac9d8a6a5ccb4029d6c2/skills/agent-architecture-audit/SKILL.md and copy the full SKILL.md text. 2. Paste it into a chat with Muse and add: "Audit this agent system: [what it does, model stack, symptoms, when it started]. Start with Phase 1 scoping." 3. Walk the evidence collection with your codebase at hand (or describe the architecture), then ask Muse to map each finding to its layer and rank by severity. Tip: answer the 7 quick diagnostic questions first. They point at the failing layer in minutes. Safety: a skill is plain-text instructions; it runs nothing by itself. Never paste secrets, credentials, or private user data into a chat, and review anything Muse proposes before it acts.
Saved to your recent installs. Find it anytime on /connect.
🔒 Security, provenance and what this touches
The install prompt below already includes the vetting steps: your agent follows the community checklist before installing anything with executable code. Want more?
Verification certificate
Dated facts about what was checked for this listing, and how. This is not a safety certification.
- Pinned version
- 4eb71d9 · 2026-10-13
- Sandbox install test
- Install ran in our test sandbox · Nov 11, 2026
- Creator identity
- Not verified
- Fingerprint watch
- Last checked · Nov 21, 2026 : no change detected in the files linked by this install prompt
- Install prompt analysis
- 2026-10-13 : No commands, network calls, packages, file writes or API keys detected.
Skill Harbor never reviews the code of a listed skill. These are dated facts about the install prompt and our install test, not a safety certification. Review a skill yourself before you install it. How we verify
Questions
How do I install a skill?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the skill up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a skill before installing it.