pm-package

pm package: resumable failure-input reduction for agents, with exact trial journals and qualified minimality receipts

pm package: constrained combinatorial test design for agents, with exact coverage receipts and resumable case outcomes

Scoped domain terms for agents: pin explicit meanings, resolve or abstain across workflows, and hand over bounded resolution packets

Beads JSONL importer extension for pm-cli

pm package: preserve units, populations and exact derivations of quantities in agent context; refuse invalid comparisons instead of mixing them

pm package: auditable evidence screening for agents, with declared criteria, per-candidate decisions and explicit unresolved candidates

Git worktree lanes for parallel agents: footprint planning from pm items, real merge-tree conflict prediction and conflict-free landing order

Agent long-term memory as typed pm items: token-budgeted recall, a prompt-ready index, claim verification and lossless import/export for Claude Code, Codex and Cursor memory

Authoring-quality lint for pm items: vague titles, unverifiable criteria, closures without evidence, stale claims — so every item is useful agent context

Full web UI for pm-cli — browse, create, update, search and manage pm projects in the browser. Self-hosted via Docker or Node.js.

GitHub Issues importer extension for pm-cli

HTTP investigations for agents: keep explicit requests and their secret-free observations as task context, compare response contracts, and hand over a bounded difference packet

Generate deterministic pm context packs for agent handoffs, reviews, and status briefs

Linear.app issue sync for pm-cli

Knowledge graph and dependency graph extension for pm CLI workspaces, with optional Neo4j sync.

Markdown TODO round-trip extension for pm-cli

Differential investigation for agents: keep competing explanations, their predictions and the observations that rule them out as tracker context, and resume an investigation from a compact packet

Complete starter/scaffold extension for pm-cli showing all capability types

Recursive Language Models on a pm workspace: typed, budgeted context reads and a resumable call ledger where every sub-call is a pm item

Evidence freshness for agents: resolve the exact source fragment an item cites, report whether it changed, and list the decisions that need re-review