PAUL is a structured development workflow for Claude Code that organizes AI-assisted work into planning, implementation, and state-reconciliation steps. It is for developers who want acceptance criteria, managed context, and completed plans while building software with Claude Code. Its catalogue entries are commands that implement the workflow.
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.
git clone --depth 1 https://github.com/ChristopherKahler/paulWrote 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/christopherkahler/paul/audit)<a href="https://agentmods.dev/commands/christopherkahler/paul/audit"><img src="https://agentmods.dev/badge/commands/christopherkahler/paul/audit/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/christopherkahler/paul/audit"><img src="https://agentmods.dev/badge/commands/christopherkahler/paul/audit.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00013 | $0.00460 |
| Opus 5.5 | $0.00005 | $0.00184 |
| Sonnet 5.5 | $0.00003 | $0.00092 |
| Haiku 4.5 | $0.00001 | $0.00046 |
Grade A, and why
paul:audit 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 1mo 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
When to use: After PLAN phase complete, before APPLY.
Optional: Only runs when explicitly invoked or suggested by enterprise_plan_audit config.
Who audits: Claude performs the audit assuming the role of senior principal engineer + compliance reviewer.
<execution_context> @~/.claude/paul-framework/workflows/audit-plan.md </execution_context>
@.paul/STATE.md @.paul/config.md
The workflow implements:
- Validate preconditions (PLAN exists, loop at PLAN complete)
- Load and parse the PLAN.md
- Execute enterprise audit (6-section structured review)
- Classify findings (must-have, strongly-recommended, can-safely-defer)
- Auto-apply must-have and strongly-recommended findings to PLAN.md
- Create AUDIT.md report in phase directory
- Update STATE.md with audit completion
- Present summary and route to APPLY
<success_criteria>
- PLAN.md audited against enterprise standards
- Must-have and strongly-recommended findings applied to PLAN.md
- AUDIT.md report created in phase directory
- STATE.md updated to reflect audit completion
- User presented with next action (run APPLY) </success_criteria>
<anti_patterns>
- Don't invent requirements not implied by the plan
- Don't assume "future phases" will fix gaps
- Don't say "this is fine for v1" unless justified
- Don't optimize for politeness over correctness
- Don't skip applying findings - the whole point is automatic remediation </anti_patterns>
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.
- 1mo ago First seen · 58 lines · 13 tokens per session scan A f2628e4c3d5a
paul:audit is a command published in the GitHub repository ChristopherKahler/paul (1,255 stars, last pushed 2mo ago), licensed MIT. It adds 13 tokens to every session and 460 once invoked, about $0.0001 per session on Opus 5.5. 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
f5-improve
Automatic code improvement and optimization.
review-artifact-comments
Works through the review comments written into a task artifact (plan or research under .ork/tasks/TASK/): lists the open ones, asks how to proceed, then replies, resolves or hands off to iterate-plan or iterate-research, and approves the artifact once nothing is open. Use when a reviewer has commented on an artifact…
ia-verify
Run pre-PR verification chain: build, types, lint, tests, security scan, diff review.
check-architecture
Angular Architecture Review.
announce
Draft X/Twitter announcement post (or thread) for the latest plugin release.
deep-audit.skeleton
Description: This section is populated by Bootstrap with manifest data. Required manifest fields: project.description, stack.primary, project.structure, project.subprojects, stack.orm, stack.authmethod Example output.