Datasets:

thomasmustier
/

superhuman-mail-sessions

Tasks:

Text Generation

Modalities:

Formats:

Languages:

Size:

Tags:

Libraries:

License:

Dataset card Data Studio Files Files and versions

Pi traces preview

Sat, Jul 11

Read /tmp/superhuman-authority-final-review-505b07c-task.md and perform that exact-head independent read-only review now. Do not modify repository files. 1 message 76 tools Read /tmp/superhuman-authority-rereview-aaadf03-task.md. Perform a supplementary independent adversarial full-diff review of exact head aaadf03f0b7339bd1b197fca77591d4c16fb5231, focusing on any remaining P0/P1 source merge blockers. Do not modify files. Write /tmp/superhuman-authority-opus-aaadf03.md and print OPUS_REVIEW_COMPLETE when done. 1 message 24 tools Read /tmp/superhuman-authority-review-task.md and perform that independent exact-head security review now. 4 messages 133 tools

Wed, Mar 25

I just saw this article. Learnings for our CLI? Anything we should look at redesigning? If you've ever watched an agent try to use a CLI, you've seen it get stuck on an interactive prompt it can't answer, or parse a help page with no examples. Most CLIs were built assuming a human is at the keyboard. Here are some things I've found that make them work better for agents: Make it non-interactive. If your CLI drops into a prompt mid-execution, an agent is stuck. It can't press arrow keys or type "y" at the right moment. Every input should be passable as a flag. Keep interactive mode as a fallback when flags are missing, not the primary path. bash # this blocks an agent $ mycli deploy ? Which environment? (use arrow keys) # this works $ mycli deploy --env staging Don't dump all your docs upfront. An agent runs mycli, sees subcommands, picks one, runs mycli deploy --help, gets what it needs. No wasted context on commands it won't use. Let it discover things as it goes. Make --help actually useful. Every subcommand gets a --help, and every --help includes examples. The examples do most of the work. An agent pattern-matches off mycli deploy --env staging --tag v1.2.3 faster than it reads a description. $ mycli deploy --help Options: --env Target environment (staging, production) --tag Image tag (default: latest) --force Skip confirmation Examples: mycli deploy --env staging mycli deploy --env production --tag v1.2.3 mycli deploy --env staging --force Accept flags and stdin for everything. Agents think in pipelines. They want to chain commands and pipe output between tools. Don't require positional args in weird orders or fall back to interactive prompts for missing values. bash cat config.json | mycli config import --stdin mycli deploy --env staging --tag $(mycli build --output tag-only) Fail fast with actionable errors. If a required flag is missing, don't hang. Error immediately and show the correct invocation. Agents are good at self-correcting when you give them something to work with. bash Error: No image tag specified. mycli deploy --env staging --tag <image-tag> Available tags: mycli build list --output tags Make commands idempotent. Agents retry constantly. Network timeouts, context getting lost mid-task. Running the same deploy twice should return "already deployed, no-op", not create a duplicate. Add --dry-run for destructive actions. Agents should be able to preview what a deploy or deletion would do before committing. Let them validate the plan, then run it for real. bash $ mycli deploy --env production --tag v1.2.3 --dry-run Would deploy v1.2.3 to production - Stop 3 running instances - Pull image registry.io/app:v1.2.3 - Start 3 new instances No changes made. $ mycli deploy --env production --tag v1.2.3 ✓ Deployed v1.2.3 to production --yes / --force to skip confirmations. Humans get "are you sure?" and agents pass --yes to bypass it. Make the safe path the default but allow bypassing. Predictable command structure. If an agent learns `mycli service list`, it should be able to guess `mycli deploy list` and `mycli config list`. Pick a pattern (resource + verb) and use it everywhere. Return data on success. Show the deploy ID and the URL. Emojis are cute, but not really needed. bash deployed v1.2.3 to staging url: https://staging.myapp.com deploy_id: dep_abc123 duration: 34s If you're building CLIs that agents will use, these patterns save a lot of debugging time. Most of the work is just making explicit what humans figured out implicitly! I bet there are many more things that I've forgotten, let me know! 3 messages 20 tools I just saw this article. Learnings for our CLI? Anything we should look at redesigning? If you've ever watched an agent try to use a CLI, you've seen it get stuck on an interactive prompt it can't answer, or parse a help page with no examples. Most CLIs were built assuming a human is at the keyboard. Here are some things I've found that make them work better for agents: Make it non-interactive. If your CLI drops into a prompt mid-execution, an agent is stuck. It can't press arrow keys or type "y" at the right moment. Every input should be passable as a flag. Keep interactive mode as a fallback when flags are missing, not the primary path. bash # this blocks an agent $ mycli deploy ? Which environment? (use arrow keys) # this works $ mycli deploy --env staging Don't dump all your docs upfront. An agent runs mycli, sees subcommands, picks one, runs mycli deploy --help, gets what it needs. No wasted context on commands it won't use. Let it discover things as it goes. Make --help actually useful. Every subcommand gets a --help, and every --help includes examples. The examples do most of the work. An agent pattern-matches off mycli deploy --env staging --tag v1.2.3 faster than it reads a description. $ mycli deploy --help Options: --env Target environment (staging, production) --tag Image tag (default: latest) --force Skip confirmation Examples: mycli deploy --env staging mycli deploy --env production --tag v1.2.3 mycli deploy --env staging --force Accept flags and stdin for everything. Agents think in pipelines. They want to chain commands and pipe output between tools. Don't require positional args in weird orders or fall back to interactive prompts for missing values. bash cat config.json | mycli config import --stdin mycli deploy --env staging --tag $(mycli build --output tag-only) Fail fast with actionable errors. If a required flag is missing, don't hang. Error immediately and show the correct invocation. Agents are good at self-correcting when you give them something to work with. bash Error: No image tag specified. mycli deploy --env staging --tag <image-tag> Available tags: mycli build list --output tags Make commands idempotent. Agents retry constantly. Network timeouts, context getting lost mid-task. Running the same deploy twice should return "already deployed, no-op", not create a duplicate. Add --dry-run for destructive actions. Agents should be able to preview what a deploy or deletion would do before committing. Let them validate the plan, then run it for real. bash $ mycli deploy --env production --tag v1.2.3 --dry-run Would deploy v1.2.3 to production - Stop 3 running instances - Pull image registry.io/app:v1.2.3 - Start 3 new instances No changes made. $ mycli deploy --env production --tag v1.2.3 ✓ Deployed v1.2.3 to production --yes / --force to skip confirmations. Humans get "are you sure?" and agents pass --yes to bypass it. Make the safe path the default but allow bypassing. Predictable command structure. If an agent learns `mycli service list`, it should be able to guess `mycli deploy list` and `mycli config list`. Pick a pattern (resource + verb) and use it everywhere. Return data on success. Show the deploy ID and the URL. Emojis are cute, but not really needed. bash deployed v1.2.3 to staging url: https://staging.myapp.com deploy_id: dep_abc123 duration: 34s If you're building CLIs that agents will use, these patterns save a lot of debugging time. Most of the work is just making explicit what humans figured out implicitly! I bet there are many more things that I've forgotten, let me know! 6 messages 107 tools

Mon, Mar 23

Currently we have the SKILL.md buried under ./skills/superhuman-mail/ Is this necessary? should we just move it to repo root so it's more discoverable (would better match how skills work anyway as you have SKILL.md at the top and then tools/other reference 4 messages 11 tools

Pi coding agent sessions for tmustier/superhuman-mail

Source project on GitHub: tmustier/superhuman-mail

This dataset contains 6 coding agent sessions recorded with pi while working on tmustier/superhuman-mail. The source project is MIT licensed.

Data description

Each *.jsonl file is one Pi session. Each line is one session entry: session headers, user and assistant messages, thinking, tool calls and results, model changes, compaction summaries and extension data.

Pi session files are tree-structured through id and parentId, so one file may contain several branches of work. See the session format documentation:

https://github.com/earendil-works/pi/blob/main/packages/coding-agent/docs/session-format.md

manifest.jsonl is the pi-share-hf index of the files in this dataset.

Redaction and review

Earlier sessions were exported with pi-share-hf, which redacts known secrets, scans with TruffleHog and uses a model review step. Sessions added or updated in October 2026 went through a stricter process:

Every session was read in full by a model reviewer looking for private information, then scanned offline with TruffleHog.
Sessions with private information that could not be removed cleanly were left out, as were trivial sessions such as greetings and smoke tests.
A human spot check followed.

Removed text is replaced with a marker such as [REDACTED_CREDENTIAL] or [Image omitted from shared copy.].

Limitations

Redaction and review reduce risk but cannot prove that no sensitive information remains. Local paths, host names and the author's GitHub identity are left in. The sessions are not a representative sample of Pi use.

Downloads last month: 28

Number of rows:

Total file size: