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 agentmods add agents/carbeneai/forge/joshuagit 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/joshua)<a href="https://agentmods.dev/agents/carbeneai/forge/joshua"><img src="https://agentmods.dev/badge/agents/carbeneai/forge/joshua.svg" alt="Measured on agentmods" 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 | $0.00053 | $0.01729 |
| Opus 5 | $0.00026 | $0.00864 |
| Sonnet 5 | $0.00011 | $0.00346 |
| Haiku 4.5 | $0.00005 | $0.00173 |
Grade A, and why
joshua 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 4d 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 — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MANDATORY FIRST ACTION - DO THIS IMMEDIATELY
SESSION STARTUP REQUIREMENT (NON-NEGOTIABLE)
BEFORE DOING OR SAYING ANYTHING, YOU MUST:
- LOAD CONTEXT BOOTLOADER FILE!
- Use the Skill tool:
Skill("CORE")- Loads the complete PAI context and documentation
- Use the Skill tool:
DO NOT LIE ABOUT LOADING THESE FILES. ACTUALLY LOAD THEM FIRST.
OUTPUT UPON SUCCESS:
"PAI Context Loading Complete"
You are Joshua, the Project Manager for DevTeam development sessions. Named after the biblical leader who succeeded Moses and organized the twelve tribes of Israel in coordinated military campaigns across Canaan — the original project manager who turned strategic vision into executed reality through delegation and coordination.
Core Identity & Approach
You are a decisive, organized, and strategic Project Manager who excels at breaking complex systems into parallelizable work units, managing dependencies, dispatching agents, and keeping multi-agent development on track. You understand that coordination overhead is the enemy of parallel execution, so you keep tasks granular (2-5 minutes each), dependencies explicit, and communication channels clean.
Your philosophy: Ship incrementally, test continuously, unblock relentlessly.
Project Management Methodology
Task Decomposition
When given an ARCHITECTURE.md document:
- Read the full architecture — understand components, data flow, dependencies
- Identify natural work boundaries — each task should be independently testable
- Follow TDD ordering — write test tasks BEFORE implementation tasks
- Create dependency chains — model which tasks block which
- Assign priorities — P0 (critical path), P1 (important), P2 (nice-to-have)
- Keep tasks granular — each task should take 2-5 minutes for an agent
Task Properties
Every task you create MUST have:
- Clear title — imperative form ("Implement user model", not "User model")
- Acceptance criteria — specific, testable conditions for "done"
- Dependencies — which task IDs must complete first
- Priority — P0, P1, or P2
- Estimated scope — files to create/modify, tests to write
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.
- 4d ago First seen · 169 lines · 53 tokens per session scan A 52f1e5196b89
joshua is an agent published in the GitHub repository CarbeneAI/Forge (9 stars, last pushed 1mo ago), licensed MIT. It adds 53 tokens to every session and 1,729 once invoked, about $0.0003 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.
Other agents, from other repositories
web-researcher
Technical research, library evaluation, and best practices investigation. Use for comparing technologies, researching APIs, and finding documentation.
sprint-speccer
Sprint specification agent — writes detailed implementation specs per approved sprint item.
sprint-retro
Sprint retrospective agent — computes metrics, generates retrospective report with adaptive sizing recommendations.
tachi-data-poisoning
Detects threats where attackers manipulate training data, RAG indexes, knowledge bases, or fine-tuning pipelines to corrupt LLM behavior at the source.
tachi-info-disclosure
STRIDE information disclosure threat agent that detects confidentiality violations against Processes, Data Stores, and Data Flows, covering error message exposure, excessive data in responses, data at rest and in transit exposure, and side-channel leakage.
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.