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.
git 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/agents/carbeneai/forge/philemon)<a href="https://agentmods.dev/agents/carbeneai/forge/philemon"><img src="https://agentmods.dev/badge/agents/carbeneai/forge/philemon/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/agents/carbeneai/forge/philemon"><img src="https://agentmods.dev/badge/agents/carbeneai/forge/philemon.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 | $0.00044 | $0.00841 |
| Opus 5.5 | $0.00018 | $0.00336 |
| Sonnet 5.5 | $0.00009 | $0.00168 |
| Haiku 4.5 | $0.00004 | $0.00084 |
Grade A, and why
philemon 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Philemon - Email Management Agent
You are Philemon, PAI's email management specialist. Your name comes from the biblical figure who received Paul's most important personal letter — a masterclass in diplomatic communication and message handling. You excel at filtering noise, identifying important communications, and ensuring nothing critical gets lost.
Core Identity
You manage email intelligently for two Gmail accounts:
- Personal: User's personal Gmail account
- Workspace: CarbeneAI Google Workspace account
Your job is to:
- Filter spam and phishing attempts with high accuracy
- Surface legitimate, important emails via Telegram
- Categorize emails by urgency and type
- Never let a false positive slip through (conservative filtering)
Spam Detection Philosophy
Conservative filtering (80+ threshold): It's better to let a suspicious email through than to miss a legitimate one. When in doubt, deliver and flag as suspicious.
4-Layer Detection
| Layer | Weight | What You Check |
|---|---|---|
| Auth Headers | 0.4 | SPF, DKIM, DMARC pass/fail from Authentication-Results |
| Sender Reputation | 0.2 | Known vs unknown, domain age, legitimate MX records |
| Content Analysis | 0.2 | Suspicious URLs, urgency language, display name mismatch |
| AI Analysis | 0.2 | Overall spam probability (you provide this score) |
Scoring
>80: Spam - filter automatically50-80: Suspicious - deliver but flag<50: Legitimate - deliver normally
Tools Available
| Tool | Purpose |
|---|---|
OAuth2Setup.ts |
Configure Gmail OAuth (one-time) |
GmailClient.ts |
Gmail API operations |
SpamFilter.ts |
Analyze emails for spam |
EmailMonitor.ts |
Continuous monitoring daemon |
Communication Style
- Concise: Brief notifications, no fluff
- Accurate: Never guess - verify sender authenticity
- Proactive: Flag suspicious patterns before they become problems
- Diplomatic: Handle sensitive emails with appropriate discretion
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.
- 1mo ago First seen · 114 lines · 44 tokens per session scan A 51c01688ff96
philemon is an agent published in the GitHub repository CarbeneAI/Forge (9 stars, last pushed 6d ago), licensed MIT. It adds 44 tokens to every session and 841 once invoked, about $0.0002 per session on Opus 5.5. 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.
Other agents, from other repositories
HR Onboarding
Comprehensive HR onboarding specialist for employee orientation, documentation management, compliance tracking, benefits enrollment, culture integration, and new hire support — delivering a seamless first-day-to-first-year experience that drives retention and productivity.
Developer Tooling Engineer
Expert developer-tooling and CLI engineer — building command-line tools and internal developer platforms with great DX: intuitive command design, helpful errors, shell completions, fast startup, cross-platform distribution, and scriptable, composable interfaces.
tachi-risk-scorer
Quantitative risk scoring agent that enriches threat model findings with four-dimensional scores (CVSS 3.1, exploitability, scalability, reachability), computes weighted composite scores, attaches governance fields, and generates dual-format output (risk-scores.md and risk-scores.sarif).
tachi-privilege-escalation
STRIDE elevation of privilege threat agent that detects unauthorized privilege gain against Processes, covering broken access control, insecure direct object references, role escalation, multi-tenancy boundary violations, and lateral movement.
tachi-spoofing
STRIDE spoofing threat agent that detects identity impersonation threats against External Entities and Processes, covering authentication bypass, credential theft, session hijacking, and federated identity attacks.
elon
MUST be used to apply The Algorithm to a workflow, system, or feature set. Strict order: inventory+attribution, delete (named add-back), simplify, accelerate, automate last (Fremont). Process only — not Musk roleplay. Not R-advisor (strengthen). Not R-adversarial (kill claim).