pandysp

Audit error handling for silent failures, inadequate feedback, and inappropriate fallbacks using one canonical methodology and the strongest isolated reviewer available.

1 skill · 1 agent

Talk to other agent sessions in the open through a shared board of plain files: post, see what's new, and find the posts that affect your task, rated by Jev in Pi with a grep fallback elsewhere.

1 skill

Surface what I think we're doing before acting on it. Produces a structured manifest exposing every silent commitment so the user can correct course before any artifact is produced.

1 skill

Surface and resolve underspecified decisions before designing. Asks targeted questions about scope, edge behavior, trade-offs, integration, constraints, and audience (with defaults), so design choices aren't silently locked around hidden assumptions.

1 skill

Classify, label or score many items, or get a fast first feeling about any question, with TypeSafe's Jev classifier from Pi's codemode.

1 skill

Deep, multi-source research on Flue workers: parallel web and optional X search, skeptical per-candidate verification and a ranked, cited report with a coverage record.

1 skill

Generate multiple strong design options with explicit tradeoff profiles, anchored against at least two ideal target architectures.

1 skill

Drive a browser with Playwright as a dual-mode superset. Resilient locators for debugging your own web app and a vision/coordinate loop for opaque real sites, on a fresh browser or the user's real logged-in session.

1 skill

Draw diagrams and sketches with Excalidraw in an Obsidian vault, and check each one with a screenshot.

1 skill

Map the terrain before designing: locate, trace, pattern, constrain. Used before any substantive design or planning step. Output is mostly for the agent's grounding; only surprises and hard constraints surface to the user.

1 skill

Author JavaScript workflows with native Flue workers, structured results, composition, retained edits, and saved-result reentry. Development draft.

1 skill

Write a durable handoff for work that's about to ship: structured PR descriptions, summaries, memos, or memory notes. Tells the next reader what was built, why, what changed, and what's next, without forcing them to dig through the diff.

1 skill

Runs a team of agent sessions in agent-manager the way a chief of staff runs an organisation: one top session, one manager per group, and sessions that do the work, with briefings, tracking and follow-up at every level.

1 skill

Explain things in plain, jargon-free language, as if the reader has no idea what you are talking about, leading with why it matters to the reader, without talking down.

1 skill

Honest self-assessment before shipping. With --fix, fix and reassess until no autofixable issues remain; keep blocked work visible.

1 skill

Systematically identify how a plan or design could fail before committing to it.

1 skill

Multi-lens quality audit for docs, code, and other artifacts. 13 quality adjectives as audit lenses with per-domain translation sheets, a workflow-backed finder/verifier pipeline, and lens-yield reporting.

1 skill

Surfaces a session's durable lessons before they fade, each with the instructions, memory, project note, or ticket it would be saved to, and writes only what the user approves.

1 skill

Get an independent review through the strongest available channel before presenting non-trivial reasoning to the user.

1 skill

Write the implementation spec (context, approach, files to modify, utilities to reuse, and verification) that drives execution after design is chosen.

1 skill

Before flipping a previously-stated position under pressure, restate its strongest form and the goal it served, then check whether the trigger refutes the core on the same goal, not just a decorative argument or a different axis.

1 skill

Turn recordings into speaker-labelled markdown notes with AssemblyAI: language detected, EU processing, paid work never repeated, notes never overwritten.

1 skill

Write code, comments, tests, or commit messages that read as if the project's own maintainer wrote them, so a contribution is accepted as native rather than rejected as foreign. A foreign-looking patch is a cost to the maintainer even when it's correct.

1 skill

Identify and verify unverified claims in the agent's own responses before presenting them as conclusions.

1 skill

Black-box verification of any agent output: code, documents, presentations, reports, configs. Examine from the consumer's perspective and show evidence.

1 skill

Route Claude Code's auto-created git worktrees through worktrunk so background and --worktree sessions inherit project hooks (env symlinks, installs, port allocation).

hooks