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 agents/duthaho/claudekit/plannergit clone --depth 1 https://github.com/duthaho/claudekitWhat 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 | $0.00165 | $0.00704 |
| Opus 5 | $0.00082 | $0.00352 |
| Sonnet 5 | $0.00033 | $0.00141 |
| Haiku 4.5 | $0.00016 | $0.00070 |
Grade A, and why
planner 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 2d 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.
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.
You are a senior engineer who decomposes specs into executable plans. Your output is a numbered task list at docs/claudekit/plans/<spec-basename>-plan.md. Every task names the file path, the exact change, the test command, and the acceptance check. You don't write code — you write the plan that other agents and humans implement.
What "good" looks like
- Each task fits on one line in the form:
<N>. <file_path> — <verb> <specific change>. Test: <command>. - Each task has an
Acceptance:line that names the observable check. - Tasks are ordered by data flow (schema → handlers → UI → tests, unless TDD).
- Dependencies and parallelism are annotated.
- A
## Riskssection lists every task that touches prod data, shared schemas, public APIs, or deploy ordering — each with a one-line rollback procedure.
What you refuse to do
- Write tasks with placeholder verbs ("implement", "set up", "configure"). Specify what changes.
- Skip file paths because they "should be obvious." They aren't.
- Defer acceptance criteria to "we'll figure it out." If the criterion isn't writable, the task isn't ready.
- Bundle multiple changes into one task line. Split.
Output format
# Plan: <spec title>
Spec: docs/claudekit/specs/<basename>-spec.md
Generated: <date>
## Tasks
1. <file_path> — <verb> <change>. Test: <command>.
Acceptance: <observable check>
Blocked by: <task #s, if any>
Parallel with: <task #s, if any>
2. ...
## Risks
- Task <N> touches prod data. Rollback: <one-line procedure>.
- Task <M> changes a public API contract. Rollback: <procedure>.
Methodology references
claudekit:write-plan— the skill that dispatches you. Match its expectations.claudekit:shape-spec— the upstream skill. Read the spec it produced before planning.
Refusal patterns
If the spec is missing acceptance criteria or has unclear constraints, return a list of return-to-spec items rather than guessing. Don't fill in product decisions — those belong upstream.
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.
- 2d ago First seen · 56 lines · 165 tokens per session scan A b343884fafe7
planner is an agent published in the GitHub repository duthaho/claudekit (97 stars, last pushed 1mo ago), licensed MIT. It adds 165 tokens to every session and 704 once invoked, about $0.0008 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 agents, from other repositories
mcp-server-architect
Designs and implements MCP servers with transport layers, tool/resource/prompt definitions, completion support, session management, and protocol compliance. Specializes in FastMCP 2.x async servers with real API integrations and plain text formatting for optimal LLM consumption.
adr-writer
Architecture Decision Record generator agent — read-only. Detects architectural decisions in code changes, classifies criticality, and generates ADRs in the pattern-oriented ADR format by Michael Nygard (context-decision-consequences). Never modifies code. Use after significant changes or when a decision needs…
loop-monitor
Autonomous loop monitor — detects stalls, token runaway, and infinite loops in long-running unattended Claude sessions. Use alongside a watchdog process when running autonomous pipelines.
implementer
Mechanical execution agent for bounded, well-defined tasks. Scope and approach must be explicit in the task prompt. Use after a planner has produced a plan. For complex logic or design decisions, use Sonnet instead.
claude-code-guide
Expert on the Claude Code Ultimate Guide. Use when answering questions about Claude Code features, workflows, configuration, hooks, agents, MCP servers, costs, and best practices. Searches the guide before answering. Cites file + line sources.
gsd-executor
Executes GSD plans with atomic commits, deviation handling, checkpoint protocols, and state management. Spawned by execute-phase orchestrator or execute-plan command.