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.
npx skills add CarbeneAI/Forge --skill paiauditgit clone --depth 1 https://github.com/CarbeneAI/ForgeWrote 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.
[](https://agentmods.dev/skills/carbeneai/forge/paiaudit)<a href="https://agentmods.dev/skills/carbeneai/forge/paiaudit"><img src="https://agentmods.dev/badge/skills/carbeneai/forge/paiaudit/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.
<a href="https://agentmods.dev/skills/carbeneai/forge/paiaudit"><img src="https://agentmods.dev/badge/skills/carbeneai/forge/paiaudit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00089 | $0.00987 |
| Opus 5 | $0.00044 | $0.00494 |
| Sonnet 5 | $0.00018 | $0.00197 |
| Haiku 4.5 | $0.00009 | $0.00099 |
Grade A, and why
PAIAudit 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 6d 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.
Reads agent configuration directorieslowAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
| **Settings** | `~/.claude/settings.json` + `settings.local.json` — hooks, permissions, env | Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
How it starts
The opening of the file, as written. The whole thing — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PAIAudit
Read-only audit of the PAI Claude Code surface. Catches risky configs, drift in symlinked external skills, and changes week-over-week.
What It Checks
| Area | Check |
|---|---|
| Settings | ~/.claude/settings.json + settings.local.json — hooks, permissions, env |
| MCP servers | ~/.claude.json — server list, command paths, token presence (count, never values) |
| Skill supply chain | All symlinks in ~/.claude/skills/ — target repo + commit SHA, drift since last run |
| Agents | ~/.claude/agents/*.md — flag broad tool access (*, Bash, Write on sensitive agents) |
| Hooks | Scripts wired in settings — existence, permissions, owner |
| File permissions | ~/.claude.json should be 0600 (holds API tokens) |
| Delta | Diff vs last audit — new skills, new MCPs, settings changes, SHA drift |
Modes
| Mode | Trigger | Output |
|---|---|---|
| Quick | audit-summary.ts |
One-line summary for Telegram daily report |
| Full | audit.ts |
Full markdown report → ~/.claude/audit-history/YYYY-MM-DD.md + Obsidian |
| Diff | audit.ts --diff |
Compare current state to previous audit, surface only changes |
Tools
| Tool | Purpose |
|---|---|
tools/audit.ts |
Full audit — markdown report with all findings |
tools/audit-summary.ts |
Quick check — single status line, exit code reflects severity |
Usage
# Full audit (saves to audit-history + prints markdown)
bun ~/.claude/skills/PAIAudit/tools/audit.ts
# Quick summary line (used by TelegramStatus)
bun ~/.claude/skills/PAIAudit/tools/audit-summary.ts
# Diff against last week's audit
bun ~/.claude/skills/PAIAudit/tools/audit.ts --diff
Severity
- CRITICAL — exposed secrets, world-writable config, broken file perms on tokens
- HIGH — unpinned external skill repos, agents with full
*access, unknown MCP servers - MEDIUM — symlink target drifted since last audit, new agents/skills added
- LOW — settings file warnings, cosmetic config issues
- INFO — counts and inventory snapshot
What ships with it
3 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.
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.
- 6d ago First seen · 90 lines · 89 tokens per session scan A a9177d96bece
PAIAudit is a skill published in the GitHub repository CarbeneAI/Forge (9 stars, last pushed 1mo ago), licensed MIT. It adds 89 tokens to every session and 987 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
scanning-containers-with-trivy-in-cicd
This skill covers integrating Aqua Security's Trivy scanner into CI/CD pipelines for comprehensive container image vulnerability detection. It addresses scanning Docker images for OS package and application dependency CVEs, detecting misconfigurations in Dockerfiles, scanning filesystem and git repositories, and…
implementing-code-signing-for-artifacts
This skill covers implementing code signing for build artifacts to ensure integrity and authenticity throughout the software supply chain. It addresses signing binaries, packages, and containers using GPG, Sigstore, and platform-specific signing tools, establishing trust chains, and verifying signatures in deployment…
implementing-semgrep-for-custom-sast-rules
Write custom Semgrep SAST rules in YAML to detect application-specific vulnerabilities, enforce coding standards, and integrate into CI/CD pipelines.
building-vulnerability-dashboard-with-defectdojo
Deploy DefectDojo as a centralized vulnerability management dashboard with scanner integrations, deduplication, metrics tracking, and Jira ticketing workflows.
panguard
AI agent security platform — audit skills, scan for threats, and run 24/7 protection with 9,700+ detection rules.
detecting-aws-credential-exposure-with-trufflehog
Detecting exposed AWS credentials in source code repositories, CI/CD pipelines, and configuration files using TruffleHog, git-secrets, and AWS-native detection mechanisms to prevent credential theft and unauthorized account access.