New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Curate Research Trace

by NeuroAIHub in NeuroAIHub/BrainPilot

Curate BrainPilot Trace Events into human-readable research Episodes, appropriately granular nodes, and direct dependson relationships.

AGPL-3.0Auto-check passed

Install Curate Research Trace

skills CLI
$ npx skills add NeuroAIHub/BrainPilot --skill curate-research-trace -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install NeuroAIHub/BrainPilot curate-research-trace --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/NeuroAIHub/BrainPilot.git skills-src && mkdir -p .claude/skills && cp -r skills-src/packages/plugin-got/skills/curate-research-trace .claude/skills/curate-research-trace && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
curate-research-trace
GitHub stars
1.1k
Token cost
~1.1k tokens
SKILL.md length
540 words
Files
3 (incl. references)
Skills in repo
59
Repo updated
First seen
Licence
AGPL-3.0

At a glance

Curate BrainPilot Trace Events into human-readable research Episodes, appropriately granular nodes, and direct dependson relationships.

  • Works in 7 steps: Inspect the active graph and reuse… → Extract units that can be inspected,… → Assign every new unit to one research… → …
  • A report contains multiple settings
  • SKILL.md covers Curate one Trace Event, Select an Episode, Decide node boundaries and Select direct parents
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Curate Research Trace is an agent skill from NeuroAIHub/BrainPilot. Curate BrainPilot Trace Events into human-readable research Episodes, appropriately granular nodes, and direct dependson relationships. Use when a report contains multiple settings, results, analyses, visualizations, findings, or conclusions; when Episode placement is unclear; or when deciding whether new evidence should create parallel/sequential nodes or update an existing node.

Its SKILL.md is about 1.1k tokens, which your agent loads only when the skill is triggered. The skill folder holds 4 other files, including reference files (for example `agents/openai.yaml` and `references/examples.md`).

The repository describes itself as: BrainPilot: Automating Brain Discovery with Agentic Research. The licence is AGPL-3.0.

When your agent uses it

  • A report contains multiple settings
  • Episode placement is unclear
  • Deciding whether new evidence should create parallel/sequential nodes
  • Update an existing node

Example prompts

  • “/curate-research-trace”

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

  1. Inspect the active graph and reuse established Episode names and nodes.
  2. Extract units that can be inspected, cited, reproduced, or revoked independently.
  3. Assign every new unit to one research work-package Episode.
  4. Update an existing node when the report adds evidence to the same unit.
  5. Create parallel nodes when neither unit consumes the other.
  6. Create sequential dependencies only when the downstream unit consumes or
  7. Add only direct parents. Never copy all transitive ancestors.

What it can do on your machine

Read from SKILL.md and the folder at commit 93f6855. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Curate Research Trace loads about 1.1k tokens when it runs, and up to ~2k if it reads all its reference files. Until then it costs about 102 tokens; SKILL.md has 540 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~102
When it runs · the whole SKILL.md, loaded when a task matches
~1.1k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from NeuroAIHub/BrainPilot at commit 93f6855, republished under its AGPL-3.0 licence (© NeuroAIHub). 540 words, ~1,085 tokens.

Download SKILL.mdSave it as .claude/skills/curate-research-trace/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.
name
curate-research-trace
description
Curate BrainPilot Trace Events into human-readable research Episodes, appropriately granular nodes, and direct depends_on relationships. Use when a report contains multiple settings, results, analyses, visualizations, findings, or conclusions; when Episode placement is unclear; or when deciding whether new evidence should create parallel/sequential nodes or update an existing node.

Curate Research Trace

Turn reported research work into a compact derivation graph. Preserve scientific units rather than tool chronology.

Curate one Trace Event

  1. Inspect the active graph and reuse established Episode names and nodes.
  2. Extract units that can be inspected, cited, reproduced, or revoked independently.
  3. Assign every new unit to one research work-package Episode.
  4. Update an existing node when the report adds evidence to the same unit.
  5. Create parallel nodes when neither unit consumes the other.
  6. Create sequential dependencies only when the downstream unit consumes or requires the upstream unit.
  7. Add only direct parents. Never copy all transitive ancestors.

Select an Episode

Treat an Episode as one coherent research subquestion or experiment family, not as a time period or causal node.

  • Use Literature Review — <topic> for evidence synthesis around one question.
  • Use Method Design — <method> for reusable method or protocol decisions.
  • Use Environment & Reproducibility for shared runtime, dependency, hardware, and reproducibility setup.
  • Use Data Preparation — <dataset> for a dataset version, split, or material preprocessing pipeline.
  • Use Main Experiment — <question> for settings and outputs testing one main hypothesis.
  • Use Ablation — <factor> for all settings, results, comparisons, visualizations, and local findings in one ablation family.
  • Use Robustness — <factor> for sensitivity or robustness settings and outputs.
  • Use Cross-experiment Analysis — <question> only when an analysis consumes results from multiple experiment families.
  • Use Final Synthesis for conclusions that integrate findings across Episodes.

Use the session language. Prefer an existing exact Episode name. Do not create Episodes based only on Agent, tool, file, timestamp, or generic lifecycle phase. Dependencies may cross Episode boundaries.

Show full SKILL.md (276 more words)Show less

Decide node boundaries

Create separate parallel nodes for:

  • distinct settings, conditions, or model variants;
  • the independently inspectable result of each setting;
  • analyses that answer different scientific questions;
  • findings that can independently hold or be refuted.

Create separate sequential nodes when one report contains independently reviewable configuration, result, analysis/visualization, finding, and conclusion units. Connect them only in the order in which outputs are actually consumed.

Update one existing node for:

  • seeds, folds, replicates, and repeated runs of the same setting and outcome;
  • additional evidence or a correction for the same result or finding;
  • multiple export formats of one result or figure;
  • immediate retries, formatting changes, and non-meaningful intermediates.

Keep multiple metrics in one result when they jointly describe one run. Split them only when they are reused independently or support independently falsifiable findings. Make a visualization a node only when it performs an independent analysis or carries an independently cited interpretation; otherwise attach it as an artifact.

Select direct parents

Apply this counterfactual: if the parent disappeared, would the child lose its evidence or need recomputation?

  • A result depends on the setting and concrete method/data/environment inputs it used.
  • An analysis depends on the result nodes it consumes.
  • A visualization depends on the result or analysis used to generate it.
  • A finding depends on its direct result or analysis evidence.
  • A conclusion depends on direct findings, not every underlying result.

Do not connect nodes because they are adjacent, share an Episode or author, use similar words, or occurred in order. Parallel settings share parents and do not depend on one another.

Read references/examples.md when a report mixes multiple research phases, contains an ablation, or leaves node boundaries unclear.

© NeuroAIHub, AGPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 2 other files (references) in packages/plugin-got/skills/curate-research-trace of NeuroAIHub/BrainPilot.

  • SKILL.md
  • agents/openai.yaml
  • references/examples.md

Open the folder on GitHubat commit 93f6855

Compare with similar skills

Curate Research Trace next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Curate Research Trace compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Curate Research Trace this skillNeuroAIHub/BrainPilot1.1k—~1.1kAutomated safety check: PassAGPL-3.0
Event Tracehoangsonww/Claude-Code-Agent-Monitor1.1k—~759Automated safety check: PassMIT
Eventscoreyhaines31/marketingskills54k—~3kAutomated safety check: PassMIT
Observe Traceruvnet/ruflo74k—~522Automated safety check: NotesMIT
Event Sourcing Architectdavila7/claude-code-templates33k4 repos~659Automated safety check: PassMIT
Event Store Designwshobson/agents40k9 repos~828Automated safety check: PassMIT

Similar skills

  • Event Trace

    hoangsonww/Claude-Code-Agent-Monitor

    Trace the full event chain for one Claude Code session into an ordered timeline of every event type with toolname and summary, highlighting gaps, out-of-order events, and failures.

    1.1k GitHub stars~759 tokensUpdated yesterday
    DevelopmentAuto-check passed

  • Events

    coreyhaines31/marketingskills

    When the user wants to plan, run, sponsor, speak at, or get pipeline from events — webinars, conferences, trade shows, meetups, dinners, workshops, virtual summits, or user conferences.

    54k GitHub stars~3k tokensUpdated 2 days ago
    Marketing & SEOAuto-check passed

  • Observe Trace

    ruvnet/ruflo

    Trace agent execution by collecting spans and building a trace tree for a task

    74k GitHub stars~522 tokensUpdated yesterday
    Auto-check: notes

  • Event Sourcing Architect

    davila7/claude-code-templates

    Expert in event sourcing, CQRS, and event-driven architecture patterns.

    33k GitHub starsUsed in 4 repos~659 tokens
    Backend & APIsAuto-check passed

  • Event Store Design

    wshobson/agents

    Designs event stores for event-sourced systems: requirements, a comparison of EventStoreDB, PostgreSQL, Kafka, DynamoDB and Marten, and stream and versioning practices.

    40k GitHub starsUsed in 9 repos~828 tokens
    Backend & APIsAuto-check passed

  • Event Delegation

    thedaviddias/Front-End-Checklist

    A skill your agent uses when reviewing scripts, client components, bundles, or runtime behavior related to Use event delegation for dynamic content.

    74k GitHub stars~500 tokensUpdated 4 days ago
    Auto-check passed

More from NeuroAIHub/BrainPilot

All 59 skills in this repo

  • Deeplabcut

    NeuroAIHub/BrainPilot

    Toolbox for markerless animal pose estimation with DeepLabCut.

    1.1k GitHub stars~1.7k tokensUpdated 8 days ago
    Auto-check passed

  • Fmriprep

    NeuroAIHub/BrainPilot

    Preprocess task-based or resting-state fMRI data with fMRIPrep — a robust, BIDS-App preprocessing pipeline built on FSL, ANTs, FreeSurfer, AFNI, and Nilearn.

    1.1k GitHub stars~4.1k tokensUpdated 8 days ago
    Auto-check passed

  • Mne Python Guide

    NeuroAIHub/BrainPilot

    Domain-validated pipeline guidance for EEG/MEG data analysis using MNE-Python: data loading, preprocessing (filtering, ICA, re-referencing), epoching, ERP/ERF computation, time-frequency…

    1.1k GitHub stars~2.3k tokensUpdated 8 days ago
    Auto-check passed

  • Netneurotools Guide

    NeuroAIHub/BrainPilot

    Domain-validated guidance for network neuroscience analysis using netneurotools: datasets, brain network metrics, connectivity consensus, modularity, spatial statistics, null models, and cortical…

    1.1k GitHub stars~2.6k tokensUpdated 8 days ago
    Auto-check passed

  • Nature Figure

    NeuroAIHub/BrainPilot

    Submission-grade Nature/high-impact journal figure workflow for Python or R.

    1.1k GitHub starsUsed in 1 repo~1.3k tokens
    Auto-check passed

  • Pycortex Guide

    NeuroAIHub/BrainPilot

    Domain-validated guidance for cortical surface visualization and brain surface rendering of fMRI data using pycortex: data types (Volume, Vertex, Dataset), 2D cortical flatmaps, 3D WebGL brain…

    1.1k GitHub stars~1.6k tokensUpdated 8 days ago
    Auto-check passed

Questions about Curate Research Trace

What does Curate Research Trace do?

Curate BrainPilot Trace Events into human-readable research Episodes, appropriately granular nodes, and direct dependson relationships. Curate Research Trace is an agent skill from NeuroAIHub/BrainPilot. Curate BrainPilot Trace Events into human-readable research Episodes, appropriately granular nodes, and direct dependson relationships.

When should I use Curate Research Trace?

Curate Research Trace fits situations like: A report contains multiple settings; episode placement is unclear; deciding whether new evidence should create parallel/sequential nodes; update an existing node.

How do I install Curate Research Trace in Claude Code?

Run `npx skills add NeuroAIHub/BrainPilot --skill curate-research-trace -a claude-code`. Or copy the skill folder (packages/plugin-got/skills/curate-research-trace in NeuroAIHub/BrainPilot) into .claude/skills/curate-research-trace in your project. Claude Code loads it when a task matches its description.

How do I install Curate Research Trace in Codex?

Run `npx skills add NeuroAIHub/BrainPilot --skill curate-research-trace -a codex`. Or copy the skill folder (packages/plugin-got/skills/curate-research-trace in NeuroAIHub/BrainPilot) into .agents/skills/curate-research-trace in your project. Codex loads it when a task matches its description.

Can I use Curate Research Trace in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add NeuroAIHub/BrainPilot --skill curate-research-trace -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/curate-research-trace, .gemini/skills/curate-research-trace, .github/skills/curate-research-trace and .opencode/skills/curate-research-trace in your project.

What does Curate Research Trace need to run?

SKILL.md names no scripts, command-line tools or credentials: Curate Research Trace is instructions for the agent only.

Does Curate Research Trace access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Curate Research Trace safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Curate Research Trace use?

Curate Research Trace is published under the AGPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Curate Research Trace use?

About 1.1k tokens (SKILL.md is roughly 4.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 926 tokens, read only when the agent opens those files.

What are the alternatives to Curate Research Trace?

Skills that share tags, products or a category with Curate Research Trace: Event Trace (hoangsonww/Claude-Code-Agent-Monitor, 1.1k stars), Events (coreyhaines31/marketingskills, 54k stars), Observe Trace (ruvnet/ruflo, 74k stars) and Event Sourcing Architect (davila7/claude-code-templates, 33k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Curate Research Trace?

NeuroAIHub (a GitHub organization) maintains it in NeuroAIHub/BrainPilot, which has 1,062 GitHub stars. The repository holds 59 skills in this directory. The repository was last updated on October 2, 2026.

Source: NeuroAIHub/BrainPilot on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.