# Toolkit prompt bank

These are the example prompts from Extending Your Agent's Toolkit, collected by task. Adapt the files, accounts, and decisions to your project. Each explanation link returns to the prompt's original location in the guide.

## Set up tools and manage project files

### Configure and check tool access

Explanation and checks: [Chapter 2: Control what added tools can access](https://iqss.github.io/dss-agentic-toolkit/chapters/tool-access.html#prompt-configure-access).

Set up access for these statistical model fits. The approved inputs are in [folder]; results and logs belong in [folder]. Leave the inputs unchanged and block access to [excluded folder]. Check the current agent settings and the queue service's access. Explain where changes are needed before applying them. Test permitted and blocked actions with harmless substitute files, then report the settings changed, the outcomes, and anything you could not verify.

### Install and check a job-queue tool

Explanation and checks: [Chapter 3: Install and manage tools with Homebrew](https://iqss.github.io/dss-agentic-toolkit/chapters/adding-tools.html#prompt-install-tools).

Set up a job queue for this project's simulations, using `pueue` unless a suitable tool is already available. Check existing installations and how software is managed on the computer that will run the jobs. Use `Homebrew` unless an established institutional method is more appropriate, particularly on a computing cluster. Install `Homebrew` if needed, and explain any steps I must complete myself. Configure the queue and confirm that you can submit and monitor jobs from this session.

### Maintain project version history

Explanation and checks: [Chapter 4: Recover earlier work: git and gh](https://iqss.github.io/dss-agentic-toolkit/chapters/version-control.html#prompt-version-history).

Maintain version history for this project using `Git`. Inspect any existing repository and preserve current edits. Track the code, manuscript, and appropriate configuration, excluding protected data, credentials, generated libraries, and caches. Record meaningful checkpoints as tasks are completed, and summarize changes affecting the analysis or reported results.

### Restore an earlier analysis decision

Explanation and checks: [Chapter 4: Recover earlier work: git and gh](https://iqss.github.io/dss-agentic-toolkit/chapters/version-control.html#prompt-recover-analysis).

Restore the earlier sample-selection rule while preserving unrelated manuscript wording and figure-formatting edits. Update the approved analysis requirements and record why the rule changed. Rerun the affected analysis, regenerate its figures, and identify manuscript values that need updating.

### Trace a variable through project files

Explanation and checks: [Chapter 5: Find and inspect project files](https://iqss.github.io/dss-agentic-toolkit/chapters/finding-files.html#prompt-search-project).

Find where `hot_days` is constructed, used in statistical models, and labeled in figures. Report the relevant file paths and line numbers. Use focused searches and excerpts. If a saved JSON or YAML response is involved, select the relevant fields rather than printing the entire response. Explain whether the construction, statistical-model use, and figure label describe the same exposure and units.

## Run and reproduce research

### Record and restore the software environment

Explanation and checks: [Chapter 6: Preserve a software environment: rig, renv, and uv](https://iqss.github.io/dss-agentic-toolkit/chapters/environments.html#prompt-restore-environment).

Inspect this project's software and record the language version and all required package versions without upgrading them. Choose a small, representative part of the analysis, run it with the current setup, and save the output. Recreate the recorded software setup in a separate environment, leaving the current one unchanged. Rerun the same part of the analysis and compare the outputs. Report anything that could not be restored and any differences in the results. Keep the setup files with the project and update the [project context files](https://iqss.github.io/dss-agentic-research/chapters/project-context.html#project-records) with any additional software requirements, restoration instructions, and the outcome of the check.

### Give research tasks saved commands

Explanation and checks: [Chapter 7: Give research tasks reliable commands: just](https://iqss.github.io/dss-agentic-toolkit/chapters/project-commands.html#prompt-saved-commands).

Inspect this project's analysis and set up named commands for fitting the statistical model, regenerating figures from saved estimates, and running the complete workflow. Use `just` unless existing project commands already cover these tasks. Check that the individual tasks can run separately. If the code needs reorganizing, save the current results first and confirm that the revised code reproduces them. Explain what each command does and use these commands whenever you carry out the corresponding tasks.

### Test the workflow before a full simulation

Explanation and checks: [Chapter 7: Give research tasks reliable commands: just](https://iqss.github.io/dss-agentic-toolkit/chapters/project-commands.html#prompt-smoke-test).

Add a `just smoke` recipe for this simulation study. Run five repetitions, fitting the planned statistical model in each and then generating the summary tables and figures. Use the same simulation and analysis code, sample size, and software environment as the full run. Save the test outputs separately from the main results. Check that each step completes and produces the expected files, and report any problems. Run this check after relevant code changes and before starting the full simulation.

### Set up selective analysis reruns

Explanation and checks: [Chapter 8: Rerun only affected steps: targets and Snakemake](https://iqss.github.io/dss-agentic-toolkit/chapters/pipeline-dependencies.html#prompt-selective-reruns).

Preserve this project's current results and software versions. Set up `targets` for this `R` analysis, installing it in the project's software environment if needed and recording its version. Use the existing analysis code to connect data preparation, the Bayesian model fits, saved posterior draws, comparison figures, and the manuscript through their dependencies. Configure the command `just all` to run the complete workflow through `targets`, and configure `just figures` to request the figures through the same runner. Show me the dependency diagram. Check selective reruns with small representative inputs, keeping test outputs separate from the study's results. Report which steps run and which results are reused, and investigate unexpected differences. Use these commands for future updates to the analysis.

### Make a simulation recoverable after interruption

Explanation and checks: [Chapter 8: Rerun only affected steps: targets and Snakemake](https://iqss.github.io/dss-agentic-toolkit/chapters/pipeline-dependencies.html#prompt-recover-interruption).

Make this simulation workflow recoverable after an interruption. Save each completed repetition so restarting does not repeat it when the inputs, code, and settings are unchanged. Check whether the fitting software can also save and reload progress within an unfinished fit, and configure that where supported. Choose how often to save progress and document its location and the restart command in the [project context files](https://iqss.github.io/dss-agentic-research/chapters/project-context.html#project-records). Test recovery on five repetitions, interrupting after some finish and while another is running. Keep test outputs separate from the study's results. Restart with the same inputs, settings, and software versions, then check that completed work is reused without missing or duplicating repetitions. Explain which unfinished work still needs to start again.

### Queue a batch of model fits

Explanation and checks: [Chapter 9: Keep unattended jobs moving: pueue](https://iqss.github.io/dss-agentic-toolkit/chapters/job-queues.html#prompt-queue-fits).

Set up `pueue` for these 100 sensitivity fits, installing it if needed and checking its background service. Make each fit run independently with the intended inputs, settings, and software versions, saving its results and log separately. Submit one representative fit through the queue, inspect its saved result, and use its memory and processor use to choose how many fits can run at once. Then submit the batch and confirm that jobs start and produce logs or outputs. Save a job list alongside the outputs with each fit's settings, queue identifier, status, result file, and log. Record how to submit and check future batches in the [project context files](https://iqss.github.io/dss-agentic-research/chapters/project-context.html#project-records), and use the queue for future unattended runs.

### Prepare a container for a collaborator

Explanation and checks: [Chapter 10: Run work in a separate environment: containers](https://iqss.github.io/dss-agentic-toolkit/chapters/containers.html#prompt-container-analysis).

Prepare a container so a collaborator can rerun this Bayesian analysis. Preserve the recorded `R` and package versions, and include the compiler and other supporting software the analysis needs. Check the intended computer and any existing container setup, then install or configure what is missing through the appropriate route. Make the required inputs read-only and save fitted results to a designated project folder outside the container. Run a small representative fit in the working setup and in the container with the same data, code, and settings. Compare the estimates and investigate unexpected differences. Check that results remain available after the test container is removed, and test required access restrictions with harmless substitute inputs. Save the configuration with the project and document how to build, run, and check it in `README.md` for collaborators. Link to these instructions from the [project context files](https://iqss.github.io/dss-agentic-research/chapters/project-context.html#project-records) and use this setup for future analysis runs.

## Build and revise documents

### Build manuscript drafts from saved results

Explanation and checks: [Chapter 11: Generate documents from research outputs: Quarto and Pandoc](https://iqss.github.io/dss-agentic-toolkit/chapters/research-documents.html#prompt-build-manuscript).

Configure this manuscript in `Quarto` to produce a `Word` draft for coauthors and a PDF for review. Check the existing document tools and install any missing components through the appropriate route. Generate the results table and insert numerical values in the text from saved analysis outputs, keeping model fitting separate from document production. Save the build as `just paper`, or use the project's existing equivalent. Build both formats and confirm that the revised sample size and group difference appear correctly in the table and accompanying text. Inspect both documents for citation, equation, figure, and layout problems, and fix any build problems. Record the setup, input files, output locations, and build instructions in the [project context files](https://iqss.github.io/dss-agentic-research/chapters/project-context.html#project-records). Use the saved command for subsequent drafts.

### Incorporate accepted coauthor edits

Explanation and checks: [Chapter 11: Generate documents from research outputs: Quarto and Pandoc](https://iqss.github.io/dss-agentic-toolkit/chapters/research-documents.html#prompt-coauthor-edits).

Preserve this coauthor's returned `Word` document and compare it with the draft sent. Inspect its comments and tracked changes, and flag conflicting edits or changes to claims and results that need my decision. Transfer the edits I have accepted into the maintained `Quarto` source. Keep unresolved comments available for review. Use the saved build command to produce a new draft without overwriting the returned copy, then check that the accepted edits appear in the new document.

## Connect to data and services

### Supply credentials to a research script

Explanation and checks: [Chapter 12: Keep credentials out of research files: 1Password and direnv](https://iqss.github.io/dss-agentic-toolkit/chapters/credentials.html#prompt-service-credentials).

Set up this project's abstract-classification script to use the chosen LLM service account within the spending limit we agreed. Check the available credential tools and configure `1Password` and `op`, or the established institutional equivalent. Guide any personal steps needed to create, store, or authorize the key without putting it in the conversation. Supply it when the script runs, and keep its value out of source files, logs, and results. Save the command, run it on one permitted abstract, and check the saved classification. Report the account used and any unresolved problem. Save the setup and sign-in instructions in the [project context files](https://iqss.github.io/dss-agentic-research/chapters/project-context.html#project-records), without the secret. Use the saved command for later runs.

### Connect to a reference library

Explanation and checks: [Chapter 13: Connect to research data and services: MCP](https://iqss.github.io/dss-agentic-toolkit/chapters/service-connections.html#prompt-connect-library).

Set up access to the `Zotero` collection "Exercise and sleep" from this agent session, using a suitable permitted integration. Check existing connections first and install or configure what is missing. Explain the account, library, and records the connection can access, including any ability to make changes. Use read-only permissions where supported and guide any personal sign-in or consent. Retrieve the paper I identify, confirm its title and identifier, and read its available attachment or note. Report what you could access and any missing material. Leave the library unchanged. Save the setup and reconnection instructions in the [project context files](https://iqss.github.io/dss-agentic-research/chapters/project-context.html#project-records), without secret values.

### Gather evidence from the connected library

Explanation and checks: [Chapter 13: Connect to research data and services: MCP](https://iqss.github.io/dss-agentic-toolkit/chapters/service-connections.html#prompt-use-library).

Use the configured connection to retrieve papers and notes from the `Zotero` collection "Exercise and sleep" for this review. Follow the agreed research question and inclusion criteria. Save an evidence table linking the relevant findings to source passages, and maintain the project's bibliography. Distinguish full texts you read from abstracts and records without accessible text. Report missing material and unresolved references. Save the outputs in the project's `literature/` folder and report their file paths. Leave the library unchanged unless I request an edit.

### Check an interactive research website

Explanation and checks: [Chapter 14: Inspect and interact with a browser](https://iqss.github.io/dss-agentic-toolkit/chapters/browser-work.html#prompt-check-website).

Start this project's local website preview and open it using the available browser tools. On the results page, select 2024 and confirm that the plotted values and labels match the saved results for that year. Switch back to 2023 and check again. Try the main navigation and inspect the pages at desktop and narrow phone-sized browser widths. Fix problems and repeat the affected checks after reloading. Report the pages, actions, and widths checked, and show any remaining issue that needs my design decision. Save the preview and checking instructions in the [project context files](https://iqss.github.io/dss-agentic-research/chapters/project-context.html#project-records) and reuse them after relevant website changes.

### Collect data from a statistics portal

Explanation and checks: [Chapter 14: Inspect and interact with a browser](https://iqss.github.io/dss-agentic-toolkit/chapters/browser-work.html#prompt-collect-web-data).

Obtain the 2024 population counts for every region from the public statistics portal I identified. Use a permitted API or export if available; inspect the website if needed to find the correct release. Confirm the year, geographic coverage, and population measure. Preserve the original download in `data/raw/` and record the source address, release, date, and selections in a source log. Check representative values and missing or duplicate regions. Create a reusable collection script with the year as an input, explicit output locations, the service's request limits, and clear failure reporting. If collection requires multiple requests, check a small retrieval before collecting the full set. Record the command and setup in the [project context files](https://iqss.github.io/dss-agentic-research/chapters/project-context.html#project-records) and use that script for future updates.

## Run and coordinate agents

### Coordinate analysis and figure agents

Explanation and checks: [Chapter 15: Divide work between agents: shared projects and git worktrees](https://iqss.github.io/dss-agentic-toolkit/chapters/parallel-work.html#prompt-coordinate-agents).

Coordinate an analysis agent and a figure agent for this project. The figure agent will write plotting code that reads results saved by the analysis agent. Before they start, specify the contents and layout of that results file, including the estimates, their units, and the lower and upper uncertainty limits. Give each agent its inputs, expected outputs, and files to edit. Let the figure agent develop against a labeled synthetic example while the analysis code is being built. Coordinate shared files and preserve existing edits. When both agents finish, combine their work and run the analysis followed by the figures using the newly produced estimates. Report any research decisions that need my input.

### Check a manuscript with a headless agent

Explanation and checks: [Chapter 16: Run agents from saved prompts](https://iqss.github.io/dss-agentic-toolkit/chapters/headless-agents.html#prompt-headless-archive).

Set up a headless check of this manuscript against its saved analysis results. Save the checking instructions and create `just check-paper`, or use the project's existing command system. Each run should start a new checking session with the selected draft, results files, and instructions needed for the check, and save its report in a new folder. Check that the command can use the intended account and tools. Have the checking agent compare reported numbers and descriptions with the saved results, identify supporting file locations, and report anything it cannot check. It should leave the manuscript and analysis unchanged. Test a small draft and a temporary copy containing a deliberate mismatch. Resolve setup failures before arranging for the check to run after each successful manuscript build. Ensure failed or incomplete checks are reported even if the manuscript builds successfully. Record how to use it in the [project context files](https://iqss.github.io/dss-agentic-research/chapters/project-context.html#project-records).

### Run interview labeling with a local LLM

Explanation and checks: [Chapter 17: Run and adapt an LLM on your own hardware](https://iqss.github.io/dss-agentic-toolkit/chapters/local-models.html#prompt-local-labeling).

Prepare the approved computer for this interview-labeling task. Check its resources, choose a compatible LLM and runtime, and configure local execution without changing my usual hosted-agent settings. Use permitted test excerpts during setup. Create a saved command that reads the labeling instructions and excerpts and saves each original response alongside the extracted identifier, category, and supporting passage. Check representative input lengths, response fields, and memory use. Preserve failed attempts and report any inputs the setup cannot process. Record the software, settings, commands, and output locations in the [project context files](https://iqss.github.io/dss-agentic-research/chapters/project-context.html#project-records).

### Prepare and evaluate an adapted LLM

Explanation and checks: [Chapter 17: Run and adapt an LLM on your own hardware](https://iqss.github.io/dss-agentic-toolkit/chapters/local-models.html#prompt-adapt-llm).

Set up the agreed adaptation study using the approved training, development, and held-out test assignments. Preserve the base LLM and the selected prompt for the unadapted comparison. Before the planned training run, check that a short run can save an adapter and reload it with the base LLM in a separate program run. Save the training configuration and progress. Evaluate the selected adapted version and the unadapted comparison on the same held-out test excerpts using the agreed criteria. Keep both sets of responses and document the commands and files needed to use the saved version again.

## Use the terminal and preserve settings

### Ask the terminal agent to regenerate a figure

Explanation and checks: [Chapter 18: Use an agent from the terminal: Claude Code](https://iqss.github.io/dss-agentic-toolkit/chapters/terminal-agents.html#prompt-terminal-figure).

Regenerate the main figure from the saved estimates and explain which files you used.

### Configure the terminal and its panes

Explanation and checks: [Appendix A: Tools for customizing your terminal](https://iqss.github.io/dss-agentic-toolkit/chapters/working-environment.html#prompt-terminal-setup).

Inspect my current terminal setup and configure `iTerm2`, `zsh`, `Oh My Zsh`, `Powerlevel10k` and `Zellij`, installing only what is missing. Preserve my existing settings and project software setup. Use a clear terminal font, a comfortable text size, and good contrast between text and background. Configure a shell prompt showing the project folder and useful status. Prepare a named session with an agent pane, a view of my project's job queue and `btop` resource monitoring. Save this arrangement as a `Zellij` layout file and show me how to reopen it. Check that a new terminal starts the intended shell, displays the prompt correctly and can launch `Claude Code` in the project. Show me how to move between panes and detach from and return to the session. Record the configuration locations and how to reopen the setup in the global user instructions file [`TOOLCHAIN.md`](https://iqss.github.io/dss-agentic-research/chapters/shared-conventions.html#specialist-conventions).

### Preserve personal settings and instructions

Explanation and checks: [Appendix B: Maintain personal settings: chezmoi](https://iqss.github.io/dss-agentic-toolkit/chapters/shared-settings.html#prompt-preserve-settings).

Preserve my [global user instructions](https://iqss.github.io/dss-agentic-research/chapters/shared-conventions.html#shared-instructions) and useful personal tool settings. Inspect their current locations and any existing settings manager before changing anything. Identify the files to manage with `chezmoi`, excluding secret values and sign-in state, and preserve their current contents. Set up the maintained source and version history, then check that applying the source produces the intended files and that the relevant tools read them. Propose a backup location outside this computer and configure it after my approval. Record how to maintain the settings, recover an earlier version, and set up another computer in the settings repository's `README.md`, linked from my [global user instructions](https://iqss.github.io/dss-agentic-research/chapters/shared-conventions.html#shared-instructions). Use that maintenance process for future settings changes.

## Diagnose a problem

### Diagnose, repair, and check a failed operation

Explanation and checks: [Appendix D: Troubleshooting](https://iqss.github.io/dss-agentic-toolkit/chapters/troubleshooting.html#prompt-diagnose-problem).

Diagnose why [operation] produced [observed result] instead of [expected result]. Inspect the exact command, relevant logs, software setup, and outputs before changing anything. Check whether an earlier attempt is still running and making progress. Explain the likely cause and make a targeted correction, preserving existing work. Use a small representative check first if the full operation is expensive, then repeat the affected operation and inspect its output. Report what changed, what now works, and anything unresolved. Record lasting setup corrections in the [project context files](https://iqss.github.io/dss-agentic-research/chapters/project-context.html#project-records), or in [global user instructions](https://iqss.github.io/dss-agentic-research/chapters/shared-conventions.html#shared-instructions) if they apply across projects. Ask me before changing research methods or inputs.
