analyze-agent-codebase

A phased architectural review of an agent-based AI codebase, meaning software where AI components perform tasks or make decisions. It indexes the repository, examines relevant parts in clusters, and combines the findings into a final analysis.

In plain words
What is it for?
Use it to document the codebase's tools, models, data entities, tests, architecture, and implementation patterns. The process runs analysis clusters sequentially using an index.
Why use it?
It provides a structured way to understand a large AI codebase without reading every file for every question. Evidence is tied to specific files, and inventories make components and tests easier to check.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/prosusai/prism/analyze-agent-codebase
Any agent
npx skills add ProsusAI/prism --skill analyze-agent-codebase
Clone the repo
git clone --depth 1 https://github.com/ProsusAI/prism

Made for: Claude Code, Codex.

Per session 32 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,475 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 $0.00032 $0.01475
Opus 5 $0.00016 $0.00737
Sonnet 5 $0.00006 $0.00295
Haiku 4.5 $0.00003 $0.00147

Measured 2d ago against content hash 7cd103f49ae5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

analyze-agent-codebase 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 2d 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.

skills/analyze-agent-codebase/SKILL.md · 145 lines

How it starts

The opening of the file, as written. The whole thing — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Analyze Agent Codebase Skill

When to use this skill

Use this skill when the user wants to perform a comprehensive architectural analysis of an agentic AI codebase.

Instructions

Analyze the target codebase in four phases.

Each cluster has its own questions file in the skill directory (e.g. questions_cluster_a.md).

Global Token Rules

  • Do NOT launch parallel subagents.
  • Run clusters sequentially.
  • Do NOT read the entire repository per cluster.
  • Read ONLY the assigned cluster questions file.
  • Read ONLY files relevant to the cluster questions.
  • Index-driven lookup only — when a section says "Find" or "Find and read", do not search the filesystem. Use the Key File Registry in index.md (already in context from Cluster A) to identify candidate files, then read only those. If a topic has no matching registry entry, write: Not indexed — verify Phase 1 captured all relevant files.

Global Output Rules

  • Iventories: always use tables (tools, models, entities, tests)
  • Findings with evidence: always use bullet points in format - [finding] — evidence: path/to/file.ext:Lxx
  • Comparisons or yes/no pattern checks: always use a table with columns: Pattern | Implemented | Evidence
  • Evidence format: path/to/file.ext:Lxx-Lyy
  • NO prose or narrative explanations (except Section 14: Decision Extraction and Classification).
  • NO restating questions.
  • NO long evidence excerpts.
  • Do NOT duplicate findings across sections.
  • Write each cluster file immediately after completion.

Per-section Output Format:

  • Provide Finding: what is implemented in the codebase.
  • Provide Evidence: exact file path and line number (file.py:Lxx) with a direct code quote.
  • If nothing is implemented, write exactly: No implementation.
  • If partially implemented, write: Partial — [what exists] / [what's missing].
  • Separate implemented-and-active, implemented-but-disabled, and not-implemented items.
  • Do not speculate. Only use code-level evidence.

Output all results to {CODEBASE_ROOT}/_analysis/ directory inside the target codebase root.

Read the full file on GitHub · 145 lines

Files

What ships with it

7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 2d ago First seen · 145 lines · 32 tokens per session scan A 7cd103f49ae5

Subscribe to this mod's changes

analyze-agent-codebase is a skill published in the GitHub repository ProsusAI/prism (20 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 32 tokens to every session and 1,475 once invoked, about $0.0002 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 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens