Ross ROSS = Recommend OSS · open-source software intelligence for agents

OthmanAdi/planning-with-files

Persistent file-based planning for AI coding agents and long-running tasks. Crash-proof markdown plans, session recovery after /clear and compaction, per-turn re-injection against context rot, deterministic completion gate. Manus-style. Install from npm, the Claude Code plugin marketplace, or npx skills. Codex, Cursor, OpenCode, 60+ agents. observed · 2026-08-28

github.com/OthmanAdi/planning-with-files · homepage · Shell · MIT (permissive) observed · 2026-08-28

Health v2 · maintenance only

78/100

  • Activity 99
  • Release rhythm 87
  • Longevity 17
How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.

  • gap_med: 1.0
  • age_days: 242
  • days_rel: 11
  • days_push: 11
  • n_releases_24m: 89

Full methodology

Adoption not part of the score

26376 stars · 2210 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

A persistent file-based planning skill/plugin for AI coding agents that maintains task_plan.md, findings.md, and progress.md on disk and re-injects them every turn. It survives context loss, /clear, crashes, and compaction, with hooks, session recovery scripts, and an opt-in completion gate, installable across 60+ agents via the Agent Skills standard.

Use cases

  • keep my coding agent's plan alive across /clear and context compaction
  • recover an agent session after a crash or context reset
  • stop my LLM agent from losing track of long multi-step tasks
  • add persistent working memory to Claude Code or Cursor
  • enforce structured planning and progress tracking for autonomous agents
  • prevent context rot during long-running agent tasks

When to choose

  • you run long or multi-step tasks with AI coding agents like Claude Code, Codex, or Cursor
  • your agent loses its plan after /clear, compaction, or crashes
  • you want Manus-style file-based working memory with per-turn re-injection
  • you want a deterministic completion gate before an agent declares a task done

When to avoid

  • you only run short single-turn prompts where persistent planning adds overhead
  • you don't want hooks that execute scripts automatically on agent events
  • your agent doesn't support the Agent Skills standard or Claude-style hooks

Facets

plugin · maturity active

agent-framework prompt-engineering workflow-automation developer-tools large-language-models developer-tools cli cross-platform claude-code agent-skills context-engineering session-recovery manus-style planning-files context-rot coding-agents hooks markdown ai-agents automation nodejs

8 sources

Member repositories

RepositoryRoleHealth v2
OthmanAdi/planning-with-filesmain78

For agents

markdown · JSON · MCP: product_card(name="OthmanAdi/planning-with-files")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem