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 skills add chrisblattman/claudeblattman --skill dr-synthesizegit clone --depth 1 https://github.com/chrisblattman/claudeblattmanWrote 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/skills/chrisblattman/claudeblattman/dr-synthesize)<a href="https://agentmods.dev/skills/chrisblattman/claudeblattman/dr-synthesize"><img src="https://agentmods.dev/badge/skills/chrisblattman/claudeblattman/dr-synthesize/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/skills/chrisblattman/claudeblattman/dr-synthesize"><img src="https://agentmods.dev/badge/skills/chrisblattman/claudeblattman/dr-synthesize.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 | $0.00022 | $0.01544 |
| Opus 5.5 | $0.00009 | $0.00618 |
| Sonnet 5.5 | $0.00004 | $0.00309 |
| Haiku 4.5 | $0.00002 | $0.00154 |
Grade A, and why
dr-synthesize 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 today.
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 — 152 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/dr-synthesize
Synthesize 2+ deep-research raw reports into a single markdown synthesis with a mandatory ## Source Reports header preserving absolute paths to originals.
Combine 2+ deep-research raw reports into a single synthesis. The synthesis is always a new file — raw inputs are never edited.
Companion to /deep-research. This skill consumes /deep-research outputs (or any deep-research raw reports with the canonical frontmatter). Called automatically by /deep-research --synthesize, but also usable standalone for ad-hoc synthesis of older reports.
Argument hint
<archived-path-1> <archived-path-2> [<archived-path-N>...] [--out <output-path>] [--topic <override>]
You need at least 2 input paths. Reports must already be archived in canonical project locations (frontmatter with date:, topic:, tool: fields).
Risk escalation: Auto for synthesis writes (new file in archive). Per-item confirm if --out overwrites an existing file.
Phase 1 — Validate inputs
For each path argument:
- Exists and is non-empty
- Has YAML frontmatter (
date,topic,toolfields present) - Lives in a project
raw-inputs/folder, e.g.:~/projects/*/raw-inputs/~/Documents/research/*/raw-inputs/~/Documents/research/*/<paper-folder>/raw-inputs/
- Legacy paths (e.g., a single central archive directory) are warned about; the central directory should hold INDEX only after federated migration.
If --topic is not provided, infer from frontmatter (must match across inputs; warn on mismatch).
Determine output project folder (Phase 3 --out default):
- If all inputs share the same parent
raw-inputs/directory → write synthesis there. - If inputs span multiple project folders → use the commissioning project (the folder of the first argument). Cross-project synthesis stubs are out of scope for v1.
Phase 2 — Build synthesis prompt
Construct an internal prompt that:
- Lists each input with its absolute path + tool + date + 1-line preview
- Asks the model to identify: agreements, contradictions, gaps, unique insights per source
- Produces a synthesis matching the schema in Phase 3
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.
- today First seen · 152 lines · 22 tokens per session scan A db9c93dd8909
dr-synthesize is a skill published in the GitHub repository chrisblattman/claudeblattman (476 stars, last pushed yesterday), licensed MIT. It adds 22 tokens to every session and 1,544 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 2026-11-06.
dr-synthesize compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| dr-synthesizechrisblattman/claudeblattman | Skill | 476 | yesterday | MIT | A | 22 | Claude Code, Codex |
| skill-builderyusufkaraaslan/Skill_Seekers | Skill | 15k | 1mo ago | MIT | A | 38 | Claude Code, Codex |
| strategic-revisionflonat/flonat-research | Skill | 149 | 3d ago | MIT | A | 66 | Claude Code |
| skill-creatorflonat/flonat-research | Skill | 149 | 3d ago | MIT | A | 41 | Claude Code |
| beamer-deckflonat/flonat-research | Skill | 149 | 3d ago | MIT | A | 63 | Claude Code |
| numerical-checkflonat/flonat-research | Skill | 149 | 3d ago | MIT | A | 65 | Claude Code |
| skill-preflightflonat/flonat-research | Skill | 149 | 3d ago | MIT | A | 34 | Claude Code |