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/piercelamb/deep-planWrote 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/agents/piercelamb/deep-plan/section-writer)<a href="https://agentmods.dev/agents/piercelamb/deep-plan/section-writer"><img src="https://agentmods.dev/badge/agents/piercelamb/deep-plan/section-writer/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/agents/piercelamb/deep-plan/section-writer"><img src="https://agentmods.dev/badge/agents/piercelamb/deep-plan/section-writer.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 | $0.00026 | $0.00268 |
| Opus 5.5 | $0.00010 | $0.00107 |
| Sonnet 5.5 | $0.00005 | $0.00054 |
| Haiku 4.5 | $0.00003 | $0.00027 |
Grade A, and why
section-writer 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 2mo 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
You are a section-writer agent for the deep-plan workflow. Your job is to generate complete, self-contained implementation section content.
Instructions
- Read the prompt file specified in the user message (format: "Read /path/to/prompt.md and execute...")
- Read all context files referenced in the prompt
- Generate the section content as specified
- Output ONLY the raw markdown content for the section
Important: A SubagentStop hook automatically extracts your output and writes it to the correct file location. You do NOT need to output JSON or specify the filename - just output the markdown content directly.
Section Content Requirements
Each section must be implementable in isolation:
- Tests FIRST (extracted from claude-plan-tdd.md).
- Implementation details (extracted from claude-plan.md)
- All necessary background context
- File paths for code to create/modify
- Dependencies on other sections (reference only, don't duplicate)
- CRITICAL Remember that tests and code should only be fully specified if absolutely necessary. Stub definitions and docstrings are fine.
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.
- 2mo ago First seen · 32 lines · 26 tokens per session scan A e8de5f448cb2
section-writer is an agent published in the GitHub repository piercelamb/deep-plan (101 stars, last pushed 4mo ago), licensed MIT. It adds 26 tokens to every session and 268 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 agents, from other repositories
spec-debugger
Fixes issues when Tester or Reviewer reject an implementation. Fresh perspective on problems the Implementer couldn't solve.
spec-scanner
Scans a codebase using LLM-driven heuristics to detect framework, patterns, entities, and registration points. Produces a persistent project profile that other agents read for wiring-aware implementation.
spec-acceptor
Performs user acceptance testing by mapping completed tasks back to requirements and verifying traceability, non-functional requirements, and overall completeness. Produces a UAT report with pass/fail per acceptance criterion and a formal sign-off recommendation. Does NOT re-run functional tests (the spec-tester…
spec-documenter
Generates user-facing documentation from spec files and implemented code. Produces API references, user guides, and architecture decision records. Context: Feature implementation is complete and user needs documentation. user: "/spec-docs" assistant: "I'll generate documentation from the spec and implementation." The…
spec-validator
Use this agent when you need to validate a spec for completeness, consistency, and implementation readiness. Examples: Context: User has finished creating a spec and wants to verify it's ready for implementation. user: "I've finished the spec for user-authentication. Can you validate it?" assistant: "I'll use the…
spec-consultant
Domain expert consultant that provides focused analysis on a specific topic during brainstorming. This is a parameterized agent — the spawning command passes the expert role, domain expertise, discussion context, and specific question via the prompt. Returns structured analysis to the Lead. Context: During…