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/christopherkahler/paul/applygit clone --depth 1 https://github.com/ChristopherKahler/paulWhat 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 | $0.00009 | $0.00493 |
| Opus 5 | $0.00005 | $0.00246 |
| Sonnet 5 | $0.00002 | $0.00099 |
| Haiku 4.5 | $0.00001 | $0.00049 |
Grade A, and why
paul:apply 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 3d 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 and plan is approved.
Executes tasks in sequence, pauses at checkpoints for user input, reports completion.
<execution_context>
@/.claude/paul-framework/workflows/apply-phase.md
@/.claude/paul-framework/references/checkpoints.md
</execution_context>
@.paul/STATE.md
Execute tasks sequentially. For each task:
- Read task definition
- Execute action
- Run verification
- Confirm done criteria
checkpoint:decision
- Present decision context and options
- Wait for user selection
- Record decision
- Continue execution
checkpoint:human-verify
- Present what was built
- Present verification steps
- Wait for "approved" or issue description
- If issues: address and re-verify
- Continue execution
checkpoint:human-action
- Present required action
- Wait for "done" confirmation
- Continue execution
<success_criteria>
- All tasks executed
- All checkpoints handled
- User informed of completion
- Next action clear (run /paul:unify) </success_criteria>
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.
- 3d ago First seen · 84 lines · 9 tokens per session scan A 7807c1e5c1b5
paul:apply is a command published in the GitHub repository ChristopherKahler/paul (1,212 stars, last pushed 12d ago), licensed MIT. It adds 9 tokens to every session and 493 once invoked, about $0.0000 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
safe-refactor
Safe refactoring with automated review, testing, and rollback capabilities.
security-review
Comprehensive security analysis with multi-layer vulnerability detection.
test
Smart test runner with filtering, coverage, and health monitoring.
implement-spec
Implement specification with full traceability and test-driven development.
refactor
Interactive refactoring assistant based on Martin Fowler's refactoring catalog.
deploy_to_docker
Build Docker image and start/redeploy the MCP Task Orchestrator container, reusing the last-used config by default.