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 commands/dormstern/forge/trygit clone --depth 1 https://github.com/dormstern/forgeWhat 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 | $0.00076 | $0.00757 |
| Opus 5 | $0.00038 | $0.00378 |
| Sonnet 5 | $0.00015 | $0.00151 |
| Haiku 4.5 | $0.00008 | $0.00076 |
Grade B, and why
try scanned grade B 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 yesterday.
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.
Strips warnings and disclaimersmediumAnti-refusal
Omitting safety caveats hides risk from the user and is a common jailbreak preamble.
Keep it tight. Don't lecture. Don't pitch the methodology. Let the kill report do the selling. How it starts
The opening of the file, as written. The whole thing — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/forge:try — The 30-second magic moment
Execution mode: direct. This command has explicit steps below — execute them in order. Do not enter plan mode, do not write a plan file, do not call ExitPlanMode. The framework already did the planning; you just run the steps.
You give a brand-new user the Forge experience in 30 seconds with zero setup. No state files. No methodology. No mode choice. Just paste-able output that sells itself.
Behavior
Step 1 — Print the pre-flight (1-2 lines)
Before showing the kill report, print:
⚡ Running the Devil's Advocate against a built-in fake startup pitch.
This is a demo — to attack your own plan, run /forge:falsify [paste plan].
This sets expectation (it's a demo, not real analysis) and previews the next command.
Step 2 — Print the kill report
Read the fixture at ${CLAUDE_PLUGIN_ROOT}/docs/demos/fixtures/try.txt and print it verbatim. Do NOT modify it, do NOT add commentary, do NOT summarize. The fixture is the entire output of this command — it represents what /forge:falsify would produce on a real plan, just pre-rendered for demo speed.
Step 3 — Print the friendly handoff (3 lines)
After the kill report, print:
✓ That was a demo on a fake plan.
Run /forge:falsify [your plan] to attack your own work
Or /forge:init when you're ready to set up Forge for your project
Keep it tight. Don't lecture. Don't pitch the methodology. Let the kill report do the selling.
Constraints
- Total runtime: ≤5 seconds. This is a magic-moment demo. If it's slow, the moment dies.
- Never modify the fixture content at runtime.
- Never dispatch the
devils-advocateagent. This is a static demo, not a real DA run. The point is instant value with zero setup. - Never ask the user any questions. The whole point is zero friction.
- If
${CLAUDE_PLUGIN_ROOT}/docs/demos/fixtures/try.txtdoesn't exist, fall back to printing: "Forge installed but demo fixture missing — please reinstall viaclaude plugin install forge@forge --force. To attack your own plan, run /forge:falsify."
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.
- yesterday First seen · 56 lines · 76 tokens per session scan B 5b7336b821bf
try is a command published in the GitHub repository dormstern/forge (6 stars, last pushed 3mo ago), licensed MIT. It adds 76 tokens to every session and 757 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it B with 1 finding (strips warnings and disclaimers). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.