Borrowing it
Nothing to install: this file belongs to jeet129/praxis. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/jeet129/praxis/main/.claude/commands/refine-idea.mdgit clone --depth 1 https://github.com/jeet129/praxisWrote 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/jeet129/praxis/refine-idea)<a href="https://agentmods.dev/commands/jeet129/praxis/refine-idea"><img src="https://agentmods.dev/badge/commands/jeet129/praxis/refine-idea.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 | $0.00044 | $0.00636 |
| Opus 5 | $0.00022 | $0.00318 |
| Sonnet 5 | $0.00009 | $0.00127 |
| Haiku 4.5 | $0.00004 | $0.00064 |
Grade A, and why
refine-idea 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.
What it actually says
Run the ideation-refinement-loop workflow for the artifact or topic named in $ARGUMENTS.
Usage examples:
/refine-idea product-brief.md/refine-idea product-brief.md --creator claude --reviewer codex/refine-idea "new onboarding concept" --creator claude --reviewer codex --enhancer claude --arbiter codex --max-passes 5 --quality-bar rigorous
Argument parsing:
- The first non-option argument is the artifact path or free-form ideation topic.
--creator <harness>binds the creator role. Default:claude.--reviewer <harness>binds the reviewer role. Default:codex.--enhancer <harness>binds the enhancer role. Default: same as creator.--arbiter <harness>binds the convergence/user-summary role. Default:codex.--max-passes <n>sets the loop bound. Default: 5.--quality-bar <standard|high|rigorous>sets the review strictness. Default:high.
Step 1 - Resolve the artifact or topic. If it is a path, read it. If it is a topic,
create the initial working artifact under .project/working/ideation/.
Step 2 - Record the harness role bindings in the run log:
creator_harness, reviewer_harness, enhancer_harness, arbiter_harness,
quality_bar, max_passes, timestamp, and source artifact/topic.
Step 3 - Invoke the workflow:
workflows/ideation-refinement-loop.yaml
Step 4 - Enforce the role separation:
- creator creates or expands the artifact.
- reviewer critiques and produces severity-tagged findings.
- enhancer evaluates the critique, accepts/rejects findings with rationale, and rewrites.
- arbiter checks convergence and prepares the user-facing approval summary.
The same harness may be bound to more than one role only if the user explicitly requests it or a default says so. Even then, keep role prompts separate and record that the same harness served multiple roles.
Step 5 - Continue until:
- no unresolved major-or-higher findings remain, and
- the next pass is expected to produce only cosmetic or low-value changes, or
--max-passesis reached.
Step 6 - Present the final result to the user with:
- final artifact path,
- role binding record,
- pass count,
- major changes made,
- accepted/rejected reviewer findings,
- remaining risks/open questions,
- recommendation: approve now or run another pass.
Do not describe the artifact as "100% perfect." Use convergence language:
ready for approval, needs another pass, or blocked on user decision.
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 · 62 lines · 44 tokens per session scan A 591948f528f3
refine-idea is a command published in the GitHub repository jeet129/praxis (7 stars, last pushed 2d ago), licensed MIT. It adds 44 tokens to every session and 636 once invoked, about $0.0002 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.
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.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.