Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add Hassaan146/forge-mentor/plugin install 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/commands/hassaan146/forge-mentor/start)<a href="https://agentmods.dev/commands/hassaan146/forge-mentor/start"><img src="https://agentmods.dev/badge/commands/hassaan146/forge-mentor/start.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 5.1 | $0.00018 | $0.04678 |
| Opus 5 | $0.00009 | $0.02339 |
| Sonnet 5 | $0.00004 | $0.00936 |
| Haiku 4.5 | $0.00002 | $0.00468 |
Grade A, and why
start 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.
How it starts
The opening of the file, as written. The whole thing — 393 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/forge:start
Turn this project into a Forge project. Run the setup interview, then begin the foundation interrogation.
Forge only acts in projects where this command has been run. Every other project stays plain Claude Code.
If this project already has notes
Call resume first. If .claude/forge/ exists, this is not a new project
and Steps 1 to 5 below are not for it. resume reads the notes and returns the
question that was on screen when the session closed, in the same shape.
- Print the banner and the update check as below. Skip the readiness check and the colour key: both were shown when this project was set up.
- Say at most one line of where things stand. Its
resumefield is that line. - Paste its
blockverbatim if there is one, and stop. That is the turn. - If
blockis empty, follow itsnext.
If it returns needs_repair, the notes are damaged: say so and run
/forge:status, which is where repair lives. Do not offer to start again: a
project's decisions are not something to re-ask for want of a header.
Nothing else happens on this path. No setup interview, no permission requests, and above all no foundation question that already has an answer: asking one twice is the fastest way to lose a user's trust in the record.
If Forge was switched off here
If .claude/forge/paused.md exists, someone ran /forge:stop in this project.
Delete it, say in one line that Forge is back on, and then take the resume path
above. The answers are already recorded.
Step 0 — is the engine even here?
Call color_legend first. It reads nothing and changes nothing, and it is
Forge's own MCP tool, so it answers one question no readiness check can: is the
engine loaded in this session.
If it is not available, the user installed Forge during this session and the engine was never started. Do this and nothing else:
python "${CLAUDE_PLUGIN_ROOT}/scripts/forge_ui.py" engine
Paste that block as your whole reply and stop. No banner, no readiness check, no explanation of how plugins load, no account of what would go wrong if you asked the questions by hand. All of that was on a real screen and none of it was the answer, which is one command long. The block already carries the two steps and the fallback.
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.
- 7d ago First seen · 393 lines · 18 tokens per session scan A ce3aa20c9587
start is a command published in the GitHub repository Hassaan146/forge-mentor (22 stars, last pushed 21d ago), licensed MIT. It adds 18 tokens to every session and 4,678 once invoked, about $0.0001 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.
Other commands, from other repositories
audit-agents-skills
Audit quality of agents, skills, and commands in a Claude Code project.
audit-whitepapers
Audit version freshness, FR/EN parity, and metadata quality of all whitepapers and recap cards.
audit-repo-docs
Audit repository documentation against 85+ best practices from claude-code-ultimate-guide.
boldguy-linkedin-answer
Draft a LinkedIn comment on a post, referencing the guide's relevant section, in Flow style with AI markers removed.
methodology-advisor
Analyzes your codebase and asks 3 targeted questions to recommend the right AI-assisted development methodology stack.
security-audit
Comprehensive security audit of your project AND Claude Code configuration. Analyzes secrets exposure, injection surfaces, dependencies, hook security, and produces a scored security posture assessment.