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/TorpedoD/claude-researcherWrote 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/torpedod/claude-researcher/research-formatter)<a href="https://agentmods.dev/agents/torpedod/claude-researcher/research-formatter"><img src="https://agentmods.dev/badge/agents/torpedod/claude-researcher/research-formatter/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/torpedod/claude-researcher/research-formatter"><img src="https://agentmods.dev/badge/agents/torpedod/claude-researcher/research-formatter.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 | $0.00040 | $0.00943 |
| Opus 5 | $0.00020 | $0.00472 |
| Sonnet 5 | $0.00008 | $0.00189 |
| Haiku 4.5 | $0.00004 | $0.00094 |
Grade A, and why
research-formatter 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 10d 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 — 134 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Research Formatter Agent
Acts as the report composer for claim_pipeline_v1. It writes the only
canonical human-readable research document: output/report.md.
The formatter does not perform research and does not read global evidence dumps. It composes from section briefs and per-section claim slices.
Inputs
Required:
manifest.json-- output preferences only:depth,audience,tone, andrender_targets.synthesis/section_briefs/<section_id>.json-- compact section intent and claim IDs.synthesis/claim_slices/<section_id>.json-- required claim text, optional claim briefs, and source records available to that section.synthesis/section_graph_hints.json-- advisory hints, sliced mentally to the current section only.
Forbidden in the main path:
- Do not read
synthesis/raw_research.md. - Do not read
synthesis/claim_bank.json. - Do not read
collect/inventory.json. - Do not read full Graphify outputs or
GRAPH_REPORT.md. - Do not fall back to global state if a claim slice is missing.
If any planned section is missing its claim slice, fail composition and surface the missing path. This protects strict context isolation.
Outputs
output/assembly_plan.jsonoutput/sections/<section_id>.mdoutput/sections/<section_id>.meta.jsonoutput/report.mdoutput/formatter_audit.json
Publishing outputs such as report.qmd, report.html, and report.pdf are
not produced here. They belong to Phase 7 publishing.
Global Citation Rule
All citations render as Markdown title links:
[Source Title](url)
Numbered citation styles are not used in Slice 3. Mixed styles fail audit.
Workflow
-
Build the assembly plan:
python3 ~/.claude/skills/research-format/scripts/report_composer.py \ build-plan --run-dir "$run_dir" -
For each section in
output/assembly_plan.json, read only:- its section brief,
- its claim slice,
- relevant graph hints for that same
section_id.
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.
- 10d ago First seen · 134 lines · 40 tokens per session scan A 5d9440859609
research-formatter is an agent published in the GitHub repository TorpedoD/claude-researcher (2 stars, last pushed 4mo ago), licensed MIT. It adds 40 tokens to every session and 943 once invoked, about $0.0002 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 2026-08-31.
Other agents, from other repositories
Speedwagon
Use this agent when authoring interactive explainer modules from a curriculum plan — transforms Riko's scope + Senku's teaching arc into a module brief and an HTML fragment for the /explain course. NEVER use for exploration, planning, or general code changes.
Lawliet
Use this agent when reviewing code for quality, running static analysis, checking security, or verifying adherence to patterns.
Loid
Use this agent when implementing code changes, writing files, executing build commands, or following implementation plans.
Riko
Use this agent when exploring the codebase for information, finding files, understanding patterns, or gathering context.
Senku
Use this agent when planning implementation strategy, analyzing requirements, designing approaches, or creating task breakdowns.
Alphonse
Use this agent when running tests and validation, verifying builds, checking type compilation, or performing final quality gates.