executing-analysis-plans

A workflow for carrying out an already approved data analysis plan. It builds the dataset, runs the specified calculations, checks alternative explanations, and assembles the results.

In plain words
What is it for?
Use it after an analysis brief or pre-analysis plan has been approved, including for robustness checks, placebo tests, alternative designs, reruns, and additional result cuts.
Why use it?
It helps keep analysis faithful to the agreed plan instead of quietly changing the question or methods during execution.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/lancegui/causal-powers/executing-analysis-plans
Any agent
npx skills add lancegui/causal-powers --skill executing-analysis-plans
Clone the repo
git clone --depth 1 https://github.com/lancegui/causal-powers

Made for: Claude Code, Codex.

Per session 177 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,272 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00177 $0.03272
Opus 5 $0.00088 $0.01636
Sonnet 5 $0.00035 $0.00654
Haiku 4.5 $0.00018 $0.00327

Measured yesterday against content hash cab6e16ffce4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

executing-analysis-plans 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 yesterday.

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.

skills/executing-analysis-plans/SKILL.md · 133 lines

How it starts

The opening of the file, as written. The whole thing — 133 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Executing Analysis Plans

Overview

A plan that's been approved is a commitment, and execution is where it either gets honored or quietly abandoned. This skill takes over once question-framing (and, for confirmatory work, pre-analysis-plan) have produced an approved plan, and carries it out: build, estimate, stress-test, assemble — including dispatching independent work to parallel subagents.

Core principle: Execute the approved plan faithfully, validating as you go and parallelizing what's independent. Autonomy here is for carrying out the agreed plan fast and thoroughly, not for changing it — any departure is a checkpoint, not a step.

Prerequisite: there is an approved plan

Don't start here from a cold "analyze this." If there's no approved brief/PAP yet, go back to question-framing (and pre-analysis-plan for confirmatory work) first — executing a plan nobody approved is just the behind-the-back problem wearing a schedule.

A new request on an already-locked plan still triggers this skill — re-fire, don't coast. A re-run, a finer reporting cut, "now do the other radii / the facility-year version" is exactly this skill's job, and verification is still worth offering before any result is written to a file (result-verification runs only if the user asks). "The design was locked last week, I'll just run it" is how a new cut ships unverified — the lock covers the design, not this run. If the new cut changes the unit or estimand, it's a question-framing/analysis-checkpoints change first, not a re-run.

The sequential spine vs. the parallel fan-out

The single biggest execution mistake is running everything in one slow serial loop — or, worse, parallelizing things that actually depend on each other. Split the plan into its dependent spine and its independent leaves.

Sequential spine (must run in order — each depends on the last):

  1. Build / clean / join the analysis dataset. Unless it's a trivial load of one already-clean file, this is a phase, not a line: delegate it to data-preparation, which validates each cleaning step (data-contracts) and routes consequential decisions to analysis-checkpoints. Nothing downstream is trustworthy until that phase's reconciliation passes.
  2. Construct the treatment, outcome, and key covariates → validate ranges, missingness, leakage.
  3. Estimate the primary specification (the one pre-committed in the PAP) → this is the number.

Read the full file on GitHub · 133 lines

Changes

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.

  1. yesterday First seen · 133 lines · 177 tokens per session scan A cab6e16ffce4

Subscribe to this mod's changes

executing-analysis-plans is a skill published in the GitHub repository lancegui/causal-powers (2 stars, last pushed 8d ago), licensed MIT. It adds 177 tokens to every session and 3,272 once invoked, about $0.0009 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.

Related

Other skills, from other repositories

audit-reproducibility

Enforce the replication-protocol.md rule by cross-checking numeric claims in a manuscript against the actual R / Stata / Python outputs. Report PASS/FAIL per claim against tolerance thresholds. Use before submission and before releasing a replication package.

pedrohcgs/claude-code-my-workflow · 54 tokens

diagnose

Root-cause a failing or wrong empirical result with a disciplined reproduce → minimise → hypothesise → instrument → fix loop, instead of guessing-and-poking. Use when the user says "why is my regression wrong", "this number changed", "my script errors out", "the result won't reproduce", "debug this", "this estimate…

pedrohcgs/claude-code-my-workflow · 145 tokens

capture-environment

Snapshot the computational environment for a replication package — detects the analysis stack (R / Stata / Python) and emits the right lockfiles (renv.lock + sessionInfo.txt, requirements.txt / environment.yml / uv.lock, Stata version + ado package list), records seeds and RNG kind, optionally writes a pinning…

pedrohcgs/claude-code-my-workflow · 139 tokens

coauthor-brief

Generate a co-author / collaborator handoff brief for a multi-author, multi-machine project — summarizing what changed since the last brief (git delta), the current state of each artifact (manuscript, analysis, slides), open questions, how to reproduce locally, and any restricted-data access steps. Use when user says…

pedrohcgs/claude-code-my-workflow · 146 tokens

data-management-plan

Draft a funder-compliant Data Management Plan (NSF DMP, NIH DMS Policy 2023, ERC, Horizon Europe) by composing the confidential-data and environment-capture primitives. Sections cover data description, formats/metadata, storage/backup, access/sharing, preservation/archiving, and roles. Use when user says "data…

pedrohcgs/claude-code-my-workflow · 143 tokens

disclosure-check

Pre-screen analysis outputs (tables, figures, logs) built on restricted or confidential data for statistical-disclosure-limitation problems before any release. Scans for small cell counts, complementary-suppression gaps, dominance (p-percent / (n,k)), re-identifiable exact counts, PII leakage, and unrounded sensitive…

pedrohcgs/claude-code-my-workflow · 147 tokens