OthmanAdi/planning-with-files is a planning system that stores an AI coding agent's task plan, findings, and progress in project files so the work can be recovered after context resets, crashes, or compaction. It is for long-running coding tasks across many agent tools, including Claude Code, Codex CLI, and Cursor. The catalogue entries are the project's commands, hooks, skills, plugin, and instruction for maintaining those planning files.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/OthmanAdi/planning-with-filesWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/othmanadi/planning-with-files/plan-loop)<a href="https://agentmods.dev/commands/othmanadi/planning-with-files/plan-loop"><img src="https://agentmods.dev/badge/commands/othmanadi/planning-with-files/plan-loop/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/othmanadi/planning-with-files/plan-loop"><img src="https://agentmods.dev/badge/commands/othmanadi/planning-with-files/plan-loop.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00040 | $0.00490 |
| Opus 5.5 | $0.00016 | $0.00196 |
| Sonnet 5.5 | $0.00008 | $0.00098 |
| Haiku 4.5 | $0.00004 | $0.00049 |
Grade A, and why
plan-loop scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Run a planning-aware cadence on top of Claude Code's /loop primitive.
Steps:
- Parse args:
- First arg matching
^\d+[smhd]$is the interval (default10m). - Remaining args are an optional task prompt.
- First arg matching
- Resolve the active plan as in
/plan-attest. - Compose the loop prompt:
- If user passed a task prompt: use it verbatim.
- Else: use the default planning tick prompt:
Read task_plan.md and progress.md. Run scripts/check-complete.sh to see remaining phases. If no progress.md entry has been added since the last loop tick, write one summarizing the current state. If a phase finished, update its Status: line in task_plan.md. Continue the next phase if work remains.
- Invoke
/loop <interval> <prompt>. - Confirm to the user: print the interval, the active plan ID, and remind that bare
/loopinvocation alone (without args) runs Claude Code's built-in maintenance prompt —/plan-loopdiffers by always grounding the tick in the planning files.
If task_plan.md does not exist, refuse and direct user to run /plan first.
Why this exists:
/loop runs prompts on cron without any plan-state contract. /plan-loop injects a plan-aware default so the recurring tick always re-reads the planning files first, runs the completion check, and writes a progress entry. Users get "babysit my plan" UX without writing a custom loop prompt.
Notes:
/plan-loopcomposes with/loop; it does not replace it./loop 5m "anything"still works.- For "babysit until plan is done" semantics: combine
/plan-loop 10m(cadence) with/plan-goal(termination criterion). The loop runs every 10 minutes; the goal stops the loop when the plan is complete. - The default tick prompt is intentionally short so it stays within compaction-safe length.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 38 lines · 40 tokens per session scan A e0d99651190e
plan-loop is a command published in the GitHub repository OthmanAdi/planning-with-files (27,535 stars, last pushed 2d ago), licensed MIT. It adds 40 tokens to every session and 490 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
design-review
Workflow recipe — review a design end-to-end, ending in measured numbers rather than adjectives, by chaining 4 skills.
replay
Summarize one Agent Monitor session by id — header plus a concise transcript recap.
firm
Convene your standing AI staff — memos on every beat, a board session without you, minutes with dissent preserved.
setup-pm-skills
Onboard a new user — find out what they do, recommend the right bundles & top skills, and set up a project CONTEXT.md so every skill is tailored to them.
promote
The promote loop — scan your transcripts for what you keep asking, pick a pattern, draft the skill, test its triggers, publish the release package.
ship-a-feature
Workflow recipe — take a feature idea from fuzzy brief to launch plan by chaining 5 skills.