ai-forge-recap

ai-forge-recap is a skill for Claude Code from robcsaszar/ai-forge. It costs 85 tokens per session (1,450 once invoked), scanned A, original, MIT.

A review tool that compares what a skill or agent claims to do with what its instructions actually make it do. It reports missing, changed, or undeclared behavior.

In plain words
What is it for?
Use it before updating a skill or agent, or when checking whether its description accurately reflects its triggers, workflow, and output.
Why use it?
Descriptions can become outdated as instructions change, making it hard to know whether an add-on still behaves as documented.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions CLAUDE.md.

Part of the robcsaszar-ai-forge plugin — 8 skills shipped together

Good fit Use it before updating a skill or agent, or when checking whether its description accurately reflects its triggers, workflow, and output.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/robcsaszar/ai-forge/ai-forge-recap
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 robcsaszar/ai-forge --skill ai-forge-recap
Clone the repo
git clone --depth 1 https://github.com/robcsaszar/ai-forge

Made for: Claude Code.

Or install robcsaszar-ai-forge, the plugin that ships this one along with the rest of its 8 skills.

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 ai-forge-recap

README.md
[![agentmods](https://agentmods.dev/badge/skills/robcsaszar/ai-forge/ai-forge-recap.svg)](https://agentmods.dev/skills/robcsaszar/ai-forge/ai-forge-recap)
Your own site
<a href="https://agentmods.dev/skills/robcsaszar/ai-forge/ai-forge-recap"><img src="https://agentmods.dev/badge/skills/robcsaszar/ai-forge/ai-forge-recap.svg" alt="Measured on agentmods" height="20"></a>
Per session 85 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,450 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 $0.00085 $0.01450
Opus 5 $0.00043 $0.00725
Sonnet 5 $0.00017 $0.00290
Haiku 4.5 $0.00009 $0.00145

Measured 7d ago against content hash 4ae97de40c22, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

ai-forge-recap 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.

skills/ai-forge-recap/SKILL.md · 143 lines

How it starts

The opening of the file, as written. The whole thing — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AI Forge Recap

Audit a skill or agent by reading its actual implementation and reporting what it really does — independent of what the description claims. The description is a claim; the body is the ground truth.


What to Read

Use the path provided or resolved by the environment. If not found, report "Not found" and stop.

For skills: also read any references/ files marked MANDATORY READ — behaviors that live outside the main body still count.

If a references/ directory exists but no files are marked MANDATORY READ, include this note in Undeclared behaviors: "references/ directory present but no MANDATORY READ triggers — possible Orphan References pattern."

For agents: read the full .agent.md file including frontmatter (model, tools, etc.).

For other file types (.instructions.md, CLAUDE.md, standalone prompts): read the full file; if the file has a leading comment or description block, treat it as the "description claim" to compare against the body.


What to Compare

The description makes four implicit claims. Check each against the body:

  1. WHAT it does — does the body do this, or something adjacent?
  2. WHEN to use it — do trigger phrases match actual entry conditions?
  3. Keywords — do searchable terms reflect what the body covers?
  4. Scope — does the description under- or over-state coverage?

Also scan for undeclared behaviors: things the body does that the description never mentions. Focus on locations where iterative edits silently accumulate new behavior: NEVER rules, phases/steps, skip/exit conditions, reference-loading triggers, confirmation loops, conditional output sections, and tool/skill invocations.

Short targets (<15 lines body): Focus on scope accuracy over exhaustive four-claim comparison. A minimal body with a matching description is Aligned — don't manufacture drift from brevity.

Before writing your verdict, ask: Would an agent picking from descriptions alone make the right call? If the description would misroute, that's Significant drift regardless of how small the wording gap appears.

Read the full file on GitHub · 143 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 · 143 lines · 85 tokens per session scan A 4ae97de40c22

Subscribe to this mod's changes

ai-forge-recap is a skill published in the GitHub repository robcsaszar/ai-forge (0 stars, last pushed 4d ago), licensed MIT. It adds 85 tokens to every session and 1,450 once invoked, about $0.0004 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 2026-08-31.

Related

Other skills, from other repositories

discount-review

Inspect the discount policy fixture with a repeatable review checklist and helper script.

promptfoo/promptfoo · 18 tokens

code-review

Reviews code for bugs, security issues, and best practices.

promptfoo/promptfoo · 11 tokens

code-review-csharp

Perform structured code reviews of C# source code covering naming conventions, performance, security, readability, and .NET best practices. Trigger phrases include "review this C# code", "check my C# for best practices", "analyze this C# class", "find issues in my C# code".

pnp/copilot-prompts · 66 tokens

mcp-server-review

Review a Model Context Protocol (MCP) server implementation against the 2026-07-28 protocol revision, which removed the initialize handshake and Mcp-Session-Id and made the protocol stateless. Use when the user asks to review, audit, upgrade or migrate an MCP server, asks whether their MCP server is spec compliant…

pnp/copilot-prompts · 93 tokens

hf-cloud-sagemaker-production-defaults

Implement a production SageMaker endpoint with autoscaling, CloudWatch alarms, and tags. Use after the serving image and IAM role are known; use the deployment planner first when architecture is undecided.

waybarrios/opencode-power-pack · 48 tokens

sarif-parsing

Parses and processes SARIF files from static analysis tools like CodeQL, Semgrep, or other scanners. Triggers on "parse sarif", "read scan results", "aggregate findings", "deduplicate alerts", or "process sarif output". Handles filtering, deduplication, format conversion, and CI/CD integration of SARIF data. Does NOT…

waybarrios/opencode-power-pack · 91 tokens