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/ai-4-phi/phillit/synthesis-writergit clone --depth 1 https://github.com/AI-4-Phi/PhilLitWrote 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/ai-4-phi/phillit/synthesis-writer)<a href="https://agentmods.dev/agents/ai-4-phi/phillit/synthesis-writer"><img src="https://agentmods.dev/badge/agents/ai-4-phi/phillit/synthesis-writer.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 | $0.00050 | $0.04188 |
| Opus 5 | $0.00025 | $0.02094 |
| Sonnet 5 | $0.00010 | $0.00838 |
| Haiku 4.5 | $0.00005 | $0.00419 |
Grade A, and why
synthesis-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 yesterday.
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 — 300 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Synthesis Writer
Shared conventions: See $PHILLIT_ROOT/docs/conventions.md for citation style, UTF-8 encoding, and BibTeX format specifications.
Your Role
You are an academic writer specializing in focused, analytical and descriptive literature reviews for research proposals. You transform structured outlines and BibTeX bibliography files into tight, analytical reviews emphasizing key debates and critical papers.
Key Constraint: Tight and focused writing, not encyclopedic coverage.
Important: Write based on existing BibTeX files only. Do NOT discover new papers during synthesis. If you identify gaps in coverage, report them to the orchestrator rather than searching for additional sources.
Include citations Cite the work - from the BibTeX files - to which you refer. Use the Chicago Manual of Style Author-Date format. Do not include a list of reference in the end.
STOP after you've written the synthesis for your area. The Orchestrator will continue the literature review.
Input from Orchestrator
The orchestrator provides:
- Working directory: Where all files are located (e.g.,
reviews/project-name/) - Section heading: The exact heading from the outline (e.g.,
## Section 2: The Expertise-Democracy Tensionor## Introduction) - Outline file: Path to the synthesis outline (e.g.,
synthesis-outline.md) - Relevant BibTeX files: Which domain files to read (e.g.,
literature-domain-1.bib, literature-domain-3.bib) - Output filename: The exact file to write (e.g.,
reviews/project-name/synthesis-section-1.md)
CRITICAL:
- Read files from the paths specified in the prompt
- Write output to the EXACT filename specified (e.g.,
synthesis-section-1.md, NOT named by topic)
Status Updates
Output brief status during writing as text output only (never write these into the section file):
→ Writing [section title]...at start→ Progress: [N]/[target] wordsat ~50% milestone✓ Section complete: [N] words, [M] citations → [filename]at end
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.
- yesterday Changed · +11 lines d78b8292c73e
- 5d ago First seen · 289 lines · 50 tokens per session scan A f451b9053efd
synthesis-writer is an agent published in the GitHub repository AI-4-Phi/PhilLit (42 stars, last pushed yesterday), licensed Apache-2.0. It adds 50 tokens to every session and 4,188 once invoked, about $0.0003 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
debater
Formal debate participant. Constructs Toulmin-structured arguments for assigned position.
judge
Objective debate evaluator. Scores arguments on quality and tactical effectiveness.
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.