Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/Jamie-BitFlight/claude_skillsnpx agentmods add skills/jamie-bitflight/claude_skills/research-curatorWrote 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/jamie-bitflight/claude_skills/research-curator)<a href="https://agentmods.dev/skills/jamie-bitflight/claude_skills/research-curator"><img src="https://agentmods.dev/badge/skills/jamie-bitflight/claude_skills/research-curator/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/jamie-bitflight/claude_skills/research-curator"><img src="https://agentmods.dev/badge/skills/jamie-bitflight/claude_skills/research-curator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Agent Snooping · line 526 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
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 | $0.00094 | $0.08062 |
| Opus 5.5 | $0.00038 | $0.03225 |
| Sonnet 5.5 | $0.00019 | $0.01612 |
| Haiku 4.5 | $0.00009 | $0.00806 |
Grade A, and why
research-curator 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 1mo 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 — 624 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Invocation Input
Read the invocation text supplied by the active harness. Parse it as the arguments described in this skill. If no text was supplied, or the host left its argument placeholder unsubstituted, ask for a supported URL or one supported mode and do not write, launch a worker, or capture a baseline.
[!IMPORTANT] Every Mermaid diagram in this file is an authoritative, executable procedure. Follow it exactly: respect sequence, conditions, branches, parallel paths, and terminal states. Do not improvise, reorder, or skip steps. If a node is ambiguous or missing required detail, pause and ask before continuing. Report the path you will follow through the diagram before executing it.
Research Curator -- Multi-Mode Orchestrator
Orchestrate research entry creation, maintenance, and validation in ./research/. Spawns @research-curator agents for content work; handles coordination, README updates, and post-actions.
Cross-Harness Role Handoff
Use the host's available-agent mechanism; do not reproduce a harness-specific tool call in the work prompt. Preserve the prompt and the worker result verbatim enough for the Agent Result Relay Rules.
| Workflow step | Claude Code route | Codex route | Required handoff |
|---|---|---|---|
| Research or repair | Select the available research-curator project agent |
Select the native research-curator role |
URL, --rerun path, or --fix path and exact issues |
| Cross-reference | Select research-cross-referencer |
Select the native research-cross-referencer role |
Add cross-references to {entry-path} |
| Insight scan | Select research-insight-extractor |
Select the native research-insight-extractor role |
Run the Overlap Scan (insight lens) on {entry-path} |
| Utilization scan | Select research-utilization-assessor |
Select the native research-utilization-assessor role |
Run the Overlap Scan (utilization lens) on {entry-path} |
The Codex roles are defined in .codex/agents/ and inherit the parent session's model, tools, permissions, and MCP configuration. Do not supply a Claude model identifier as a Codex role model. If the named role is unavailable in the active host, stop before mutation and report that exact unavailable route; do not substitute a generic worker.
What ships with it
16 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/batch-mode.md 4.9 KB
- references/cross-reference-format.md 4.4 KB
- references/duplicate-detection.md 1.1 KB
- references/entry-quality-standards.md 6.9 KB
- references/entry-review-rubric.md 15 KB
- references/entry-template.md 5.4 KB
- references/extraction-methodology.md 7.9 KB
- references/overlap-scan.md 2.3 KB
- references/repo-access-procedure.md 4.5 KB
- references/validation-rules.md 5.8 KB
- scripts/backlink_cache.py 7.0 KB runs code
- scripts/backlink_lib.py 34 KB runs code
- scripts/backlink_models.py 2.2 KB runs code
- scripts/fix_research_formatting.py 13 KB runs code
- scripts/validate_research.py 43 KB runs code
- SKILL-GOALS.md 1.5 KB
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.
- 1mo ago Changed · +43 lines 202aa5339f2c
- 1mo ago Changed · +57 lines 3e4723082490
- 1mo ago Changed · -2 lines 970614b0da1c
- 2mo ago Changed a336cae58679
- 2mo ago Changed · +11 lines 2d803b711dbc
- 2mo ago Changed · -33 lines dbb40e1afe76
- 2mo ago Changed · +8 lines 12a31a13fdc2
- 2mo ago Changed · +13 lines 20a5d505efcc
- 2mo ago First seen · 527 lines · 94 tokens per session scan A ef369484c459
research-curator is a skill published in the GitHub repository Jamie-BitFlight/claude_skills (67 stars, last pushed 29d ago), licensed MIT. It adds 94 tokens to every session and 8,062 once invoked, about $0.0004 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 2026-09-03.
research-curator compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| research-curatorJamie-BitFlight/claude_skills | Skill | 67 | 29d ago | MIT | A | 94 | Claude Code |
| imagegenopenai/codex | Skill | 136k | today | Apache-2.0 | A | 113 | Codex |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 80k | yesterday | MIT | A | 200 | Claude Code, Codex |
| playwright-climicrosoft/playwright | Skill | 100k | today | Apache-2.0 | A | 19 | Claude Code |
| browser-automationopenclaw/openclaw | Skill | 395k | today | MIT | A | 36 | Claude Code |
| free-llm-apismnfst/awesome-free-llm-apis | Skill | 14k | today | CC0-1.0 | A | 113 | Claude Code, Codex |
| next-cache-components-adoptionvercel/next.js | Skill | 144k | today | MIT | A | 95 | Claude Code |