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/skulitom/agentui/primegit clone --depth 1 https://github.com/skulitom/AgentUIWhat 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 | $0.00016 | $0.00993 |
| Opus 5 | $0.00008 | $0.00496 |
| Sonnet 5 | $0.00003 | $0.00199 |
| Haiku 4.5 | $0.00002 | $0.00099 |
Grade A, and why
prime 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 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.
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 — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Delegate this task to Prime Agent and report the result: $ARGUMENTS
What Prime Agent is
A separate agent CLI (prime-agent, v0.7.0, installed globally via npm). It drives a
persistent IPython kernel rather than discrete file tools — every action it takes is
Python it wrote. It is not a Claude Code subagent: it has its own model, its own
provider auth, and its own session store in ~/.prime/agent/sessions/.
It discovers CLAUDE.md natively by walking up from its cwd, so it already sees this
repo's conventions. Do not paste CLAUDE.md into the prompt.
Required environment (Windows)
Prime Agent 0.7.0 auto-bootstraps its kernel at kernel-venv/bin/python — the POSIX
layout. On Windows the interpreter is at kernel-venv/Scripts/python.exe, so the
bootstrap fails, and a failed bootstrap recreates the venv and wipes what is in it.
Every invocation below must export the override first, or the agent runs with no tools
at all and can only answer from context:
export PRIME_AGENT_KERNEL_PYTHON='C:\Users\artem\.prime\agent\kernel-venv\Scripts\python.exe'
That venv is provisioned with ipykernel, prime-agent-runtime, and the default set
(requests httpx pyyaml tomli python-dotenv pandas numpy scipy beautifulsoup4 lxml pydantic). If the agent reports missing packages, install them into that same
interpreter with uv pip install --python <path> <pkgs> — never let the broken
auto-bootstrap try, since it clears the venv on failure.
Invoking it
One-shot, prints the final response and exits:
export PRIME_AGENT_KERNEL_PYTHON='C:\Users\artem\.prime\agent\kernel-venv\Scripts\python.exe'; prime-agent -p --cwd "C:/DEV/AgentUI" "<task>"
Long or open-ended work — run it in the background and read the log, rather than blocking a foreground Bash call for minutes:
export PRIME_AGENT_KERNEL_PYTHON='C:\Users\artem\.prime\agent\kernel-venv\Scripts\python.exe'; prime-agent -p --cwd "C:/DEV/AgentUI" "<task>" > .prime-run.log 2>&1
Bounded autonomous run, when the task needs many turns to converge. Every limit is a hard stop, and hitting one does NOT mean the task succeeded — check the output:
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 · 79 lines · 16 tokens per session scan A f81d0f71b21a
prime is a command published in the GitHub repository skulitom/AgentUI (0 stars, last pushed 26d ago), licensed MIT. It adds 16 tokens to every session and 993 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 2026-08-31.
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.