Agent skill

Run Phase

by xinzhuwang-wxz in xinzhuwang-wxz/OpenPE

Execute a single phase of the analysis pipeline by phase identifier

GPL-3.0Auto-check passed

Install Run Phase

skills CLI

$ npx skills add xinzhuwang-wxz/OpenPE --skill run-phase -a claude-code

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

GitHub CLI

$ gh skill install xinzhuwang-wxz/OpenPE run-phase --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/xinzhuwang-wxz/OpenPE.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/run-phase .claude/skills/run-phase && 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/

Paste into Claude Code

Install the "run-phase" agent skill from https://github.com/xinzhuwang-wxz/OpenPE/tree/master/.claude/skills/run-phase into .claude/skills/run-phase/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "run-phase", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

$skill-installer

$skill-installer install https://github.com/xinzhuwang-wxz/OpenPE/tree/master/.claude/skills/run-phase

Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.

skills CLI

$ npx skills add xinzhuwang-wxz/OpenPE --skill run-phase -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install xinzhuwang-wxz/OpenPE run-phase --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/xinzhuwang-wxz/OpenPE.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.claude/skills/run-phase .agents/skills/run-phase && rm -rf skills-src

Use ~/.agents/skills/ instead of .agents/skills for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "run-phase" agent skill from https://github.com/xinzhuwang-wxz/OpenPE/tree/master/.claude/skills/run-phase into .agents/skills/run-phase/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "run-phase", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add xinzhuwang-wxz/OpenPE --skill run-phase -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install xinzhuwang-wxz/OpenPE run-phase --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/xinzhuwang-wxz/OpenPE.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.claude/skills/run-phase .cursor/skills/run-phase && rm -rf skills-src

Use ~/.cursor/skills/ instead of .cursor/skills for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "run-phase" agent skill from https://github.com/xinzhuwang-wxz/OpenPE/tree/master/.claude/skills/run-phase into .cursor/skills/run-phase/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "run-phase", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

gemini skills install

$ gemini skills install https://github.com/xinzhuwang-wxz/OpenPE.git --path .claude/skills/run-phase

--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.

skills CLI

$ npx skills add xinzhuwang-wxz/OpenPE --skill run-phase -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install xinzhuwang-wxz/OpenPE run-phase --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/xinzhuwang-wxz/OpenPE.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.claude/skills/run-phase .gemini/skills/run-phase && rm -rf skills-src

Use ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "run-phase" agent skill from https://github.com/xinzhuwang-wxz/OpenPE/tree/master/.claude/skills/run-phase into .gemini/skills/run-phase/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "run-phase", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install xinzhuwang-wxz/OpenPE run-phase

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add xinzhuwang-wxz/OpenPE --skill run-phase -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

Manual copy

$ git clone --depth 1 https://github.com/xinzhuwang-wxz/OpenPE.git skills-src && mkdir -p .github/skills && cp -r skills-src/.claude/skills/run-phase .github/skills/run-phase && rm -rf skills-src

Use ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "run-phase" agent skill from https://github.com/xinzhuwang-wxz/OpenPE/tree/master/.claude/skills/run-phase into .github/skills/run-phase/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "run-phase", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add xinzhuwang-wxz/OpenPE --skill run-phase -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install xinzhuwang-wxz/OpenPE run-phase --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/xinzhuwang-wxz/OpenPE.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.claude/skills/run-phase .opencode/skills/run-phase && rm -rf skills-src

Use ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "run-phase" agent skill from https://github.com/xinzhuwang-wxz/OpenPE/tree/master/.claude/skills/run-phase into .opencode/skills/run-phase/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "run-phase", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: run-phase
GitHub stars: 101
Token cost: ~1.7k tokens
SKILL.md length: 787 words
Files: 1
Skills in repo: 5
Repo updated: Apr 2, 2026
First seen: Oct 7, 2026
Licence: GPL-3.0

At a glance

Execute a single phase of the analysis pipeline by phase identifier

Works in 6 steps: Read Current State → Read Phase Methodology → Identify Upstream Artifacts → …
SKILL.md covers Step 1: Read Current State, Step 2: Read Phase Methodology, Step 3: Identify Upstream… and Step 4: Execute the Phase, plus 2 more sections
Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Run Phase is an agent skill from xinzhuwang-wxz/OpenPE. Execute a single phase of the analysis pipeline by phase identifier

Its SKILL.md is about 1.7k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

The licence is GPL-3.0.

Example prompts

“/run-phase”

Workflow steps

6 steps, taken from the step headings in SKILL.md.

Read Current State
Read Phase Methodology
Identify Upstream Artifacts
Execute the Phase
Run Review
Update State and Report

What it can do on your machine

Read from SKILL.md and the folder at commit f29b438. 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

Run Phase loads about 1.7k tokens when it runs. Until then it costs about 19 tokens; SKILL.md has 787 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~19
When it runs · the whole SKILL.md, loaded when a task matches: ~1.7k

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 xinzhuwang-wxz/OpenPE at commit f29b438, republished under its GPL-3.0 licence (© xinzhuwang-wxz). 787 words, ~1,704 tokens.

Download SKILL.mdSave it as .claude/skills/run-phase/SKILL.md (or your agent's skills folder).

name: run-phase
description: Execute a single phase of the analysis pipeline by phase identifier
user-invocable: true

/run-phase -- Execute a Single Analysis Phase

Run a single phase of the analysis pipeline.

Arguments: $ARGUMENTS

The argument is a phase identifier: 1, 2, 3, 4a, 4b, 4c, or 5.

Step 1: Read Current State

Find the analysis directory (look for STATE.md in the current directory or immediate subdirectories, including under analyses/).
Read STATE.md to confirm the analysis is at the correct phase. If the requested phase does not match the current phase in STATE.md, report the mismatch and stop unless the user explicitly confirms they want to run this phase out of order.
Read analysis_config.yaml for configuration (model tier, channels, cost controls, pixi settings).

Step 2: Read Phase Methodology

Read src/methodology/03-phases.md and locate the section for the requested phase. Also read:

orchestration/agents.md -- agent definitions
orchestration/sessions.md -- session naming and isolation
CLAUDE.md -- orchestration model

Read the conventions/ symlink in the analysis directory to identify applicable coding, plotting, and naming conventions. All spawned agents must follow these conventions.

Step 3: Identify Upstream Artifacts

Each phase depends on artifacts from prior phases. Locate the latest version of each required artifact by finding the most recent file matching the artifact pattern (sorted by timestamp in the filename).

Phase	Required upstream artifacts
1	`prompt.md` only
2	`prompt.md`, `phase1_strategy/exec/STRATEGY*.md` (latest)
3	`prompt.md`, `STRATEGY.md`, `phase2_exploration/exec/EXPLORATION.md`
4a	`prompt.md`, `STRATEGY.md`, `SELECTION.md` (per channel or combined)
4b	`prompt.md`, `STRATEGY.md`, `SELECTION.md`, `phase4_inference/4a_expected/exec/INFERENCE_EXPECTED*.md`
4c	All of 4b's inputs plus `phase4_inference/4b_partial/exec/INFERENCE_PARTIAL*.md`, and confirmed `approved_for_verification: true` in config
5	All prior phase artifacts

Step 4: Execute the Phase

Update STATE.md: status: executing, current phase: {phase}, timestamp.

All agents use pixi run for script execution and must read applicable conventions/ files.

Phase 1: Strategy

Spawn lead-analyst via SendMessage:

Inputs: prompt.md, methodology (Phase 1 section), analysis_config.yaml
Must read: applicable conventions/ files for naming and coding standards
Working directory: phase1_strategy/
Expected output: exec/STRATEGY.md, updates to experiment_log.md
The agent should query the data corpus, identify signal/backgrounds, propose selection, define verification strategy, outline systematics
All scripts use pixi run for execution

Phase 2: Exploration

Spawn three agents in parallel via SendMessage:

data-explorer: sample inventory, data quality checks, baseline yields
domain-specialist: variable definitions, data validation
domain-scout: expected relationships, domain predictions, prior results

All read: prompt.md, latest STRATEGY.md, methodology (Phase 2 section) All read: applicable conventions/ files All write to: phase2_exploration/ All use pixi run for script execution

After all complete, spawn lead-analyst to consolidate into exec/EXPLORATION.md.

Phase 3: Selection and Background Modeling

Check analysis_config.yaml for channels. For each channel (or the single analysis):

Spawn signal-lead and background-estimator in parallel
Inputs: prompt.md, STRATEGY.md, EXPLORATION.md, methodology (Phase 3 section)
Must read: applicable conventions/ files
Output: exec/SELECTION.md (or per-channel SELECTION_{CHANNEL}.md)
All scripts use pixi run for execution

If multi-channel, after all channels complete, spawn lead-analyst to produce SELECTION_COMBINED.md.

Show full SKILL.md (339 more words)Show less

Phase 4a: Expected Results

Identify systematic sources from the strategy and selection artifacts
Spawn systematic-source-evaluator agents in parallel (one per source)
- Must read: applicable conventions/ files
- All scripts use pixi run for execution
After all complete, spawn systematics-fitter:
- Builds statistical model, runs Asimov fits, signal injection tests
- Output: exec/INFERENCE_EXPECTED.md

Phase 5: Partial Verification

Spawn systematics-fitter:
- Runs fit on 10% SR data subsample (fixed random seed)
- Output: exec/INFERENCE_PARTIAL.md
- Uses pixi run for execution
Spawn note-writer:
- Produces exec/ANALYSIS_NOTE_DRAFT.md and exec/VERIFICATION_CHECKLIST.md
- Must read: applicable conventions/ files for document formatting

Phase 6: Full Verification

Verify analysis_config.yaml has approved_for_verification: true. If not, STOP and instruct the user to run /approve-verification first.
Spawn systematics-fitter:
- Runs full fit on complete dataset
- Output: exec/INFERENCE_OBSERVED.md
- Uses pixi run for execution
Spawn cross-checker:
- Validates consistency with partial and expected results

Phase 5: Documentation

Spawn note-writer:

Reads all phase artifacts including ANALYSIS_NOTE_DRAFT.md and INFERENCE_OBSERVED.md
Must read: applicable conventions/ files for document and figure formatting
Output: exec/ANALYSIS_NOTE.md

Step 5: Run Review

After execution completes, update STATE.md: status: reviewing.

Invoke the review by running /review-phase with the current phase number, OR run the review inline:

Phase	Review tier
1	4-bot (physics + critical + constructive in parallel, then arbiter) with plot-validator
2	Self-review (no separate review -- already done by executor)
3	1-bot per channel (critical reviewer) with plot-validator
4a	4-bot with plot-validator
4b	4-bot, then PAUSE for human gate, with plot-validator
4c	1-bot with plot-validator
5	5-bot (physics + critical + constructive + rendering in parallel, then arbiter) with plot-validator

Step 6: Update State and Report

On review PASS:

Update STATE.md: status: passed for this phase, add row to Phase History table
For Phase 4b specifically: set status: human_gate and PAUSE
Report: which phase completed, artifact location, review outcome, iteration count

On review ITERATE:

The review command handles re-spawning the executor. Wait for it to complete.

On review ESCALATE:

Update STATE.md: status: blocked
Report the escalation reason to the user

Check for regression triggers in review output. If found, handle per the regression protocol in CLAUDE.md.

Files

Just SKILL.md in .claude/skills/run-phase of xinzhuwang-wxz/OpenPE.

Open the folder on GitHubat commit f29b438

Compare with similar skills

Run Phase 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.

Run Phase compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Run Phase this skillxinzhuwang-wxz/OpenPE	101	—	~1.7k	Automated safety check: Pass	GPL-3.0	6 mo ago
Gsd Execute Phaseopen-gsd/gsd-core	10k	1 repos	~801	Automated safety check: Notes	MIT	yesterday
Executealirezarezvani/claude-skills	28k	—	~831	Automated safety check: Pass	MIT	1 mo ago
Debugging Executionsn8n-io/n8n	207k	—	~2.6k	Automated safety check: Pass	Custom licence	today
Ulw Executecode-yeongyu/oh-my-openagent	70k	—	~6.3k	Automated safety check: Pass	Custom licence	today
Executebrycewang-stanford/Auto-Empirical-Research-Skills	4.6k	—	~370	Automated safety check: Notes	Custom licence	5 days ago

Similar skills

Gsd Execute Phase
open-gsd/gsd-core
SDD phase execution — execute all plans in a phase with dependency-aware wave parallelization
10k GitHub starsUsed in 1 repo~801 tokens
Agent WorkflowsAuto-check: notes
Execute
alirezarezvani/claude-skills
/cs:execute <decision — Generate a 90-day execution plan with weekly milestones, DRIs, and check-in cadence from an approved decision.
28k GitHub stars~831 tokensUpdated 1 mo ago
Product & Project ManagementAuto-check passed
Debugging Executions
n8n-io/n8n
Official
Debug failed or wrong-output workflow executions using executions tools.
207k GitHub stars~2.6k tokensUpdated today
DevelopmentAuto-check passed
Ulw Execute
code-yeongyu/oh-my-openagent
Executes a written ulw-plan work plan with Boulder state, evidence ledger, worktree discipline, and parallel subagents.
70k GitHub stars~6.3k tokensUpdated today
Agent WorkflowsAuto-check passed
Execute
brycewang-stanford/Auto-Empirical-Research-Skills
Executes all registered notebooks, strips noisy cell metadata, and syncs Jupytext pairs.
4.6k GitHub stars~370 tokensUpdated 5 days ago
Data & AnalyticsAuto-check: notes
Gsd Phase
open-gsd/gsd-core
Multi-phase management — add, insert, remove, or edit phases in ROADMAP.md (roadmap phase CRUD)
10k GitHub starsUsed in 1 repo~603 tokens
Product & Project ManagementAuto-check: notes

Questions about Run Phase

What does Run Phase do?

Execute a single phase of the analysis pipeline by phase identifier. Run Phase is an agent skill from xinzhuwang-wxz/OpenPE.

How do I install Run Phase in Claude Code?

Run `npx skills add xinzhuwang-wxz/OpenPE --skill run-phase -a claude-code`. Or copy the skill folder (.claude/skills/run-phase in xinzhuwang-wxz/OpenPE) into .claude/skills/run-phase in your project. Claude Code loads it when a task matches its description.

How do I install Run Phase in Codex?

Run `npx skills add xinzhuwang-wxz/OpenPE --skill run-phase -a codex`. Or copy the skill folder (.claude/skills/run-phase in xinzhuwang-wxz/OpenPE) into .agents/skills/run-phase in your project. Codex loads it when a task matches its description.

Can I use Run Phase 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 xinzhuwang-wxz/OpenPE --skill run-phase -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/run-phase, .gemini/skills/run-phase, .github/skills/run-phase and .opencode/skills/run-phase in your project.

What does Run Phase need to run?

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

Does Run Phase 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 Run Phase 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 Run Phase use?

Run Phase is published under the GPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Run Phase use?

About 1.7k tokens (SKILL.md is roughly 6.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Run Phase?

Skills that share tags, products or a category with Run Phase: Gsd Execute Phase (open-gsd/gsd-core, 10k stars), Execute (alirezarezvani/claude-skills, 28k stars), Debugging Executions (n8n-io/n8n, 207k stars) and Ulw Execute (code-yeongyu/oh-my-openagent, 70k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Run Phase?

xinzhuwang-wxz (a GitHub user) maintains it in xinzhuwang-wxz/OpenPE, which has 101 GitHub stars. The repository holds 5 skills in this directory. The repository was last updated on April 2, 2026.

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

Install Run Phase

Facts

At a glance

What it does

Example prompts

Workflow steps

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

/run-phase -- Execute a Single Analysis Phase

Step 1: Read Current State

Step 2: Read Phase Methodology

Step 3: Identify Upstream Artifacts

Step 4: Execute the Phase

Phase 1: Strategy

Phase 2: Exploration

Phase 3: Selection and Background Modeling

Phase 4a: Expected Results

Phase 5: Partial Verification

Phase 6: Full Verification

Phase 5: Documentation

Step 5: Run Review

Step 6: Update State and Report

Files

Compare with similar skills

Similar skills

More from xinzhuwang-wxz/OpenPE

Questions about Run Phase