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.
npx agentmods add skills/nerds-odd-e/doughnut/execute-plannpx skills add nerds-odd-e/doughnut --skill execute-plangit clone --depth 1 https://github.com/nerds-odd-e/doughnutWhat 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 | $0.00081 | $0.03144 |
| Opus 5 | $0.00041 | $0.01572 |
| Sonnet 5 | $0.00016 | $0.00629 |
| Haiku 4.5 | $0.00008 | $0.00314 |
Grade A, and why
execute-plan 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 3d 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.
How it starts
The opening of the file, as written. The whole thing — 270 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Purpose: Local execution overlay for GSD plans — complements
/gsd-execute-phase but requires this repo's wrap-up per slice per
.cursor/rules/gsd-coexistence.mdc.
Output: Slices completed with commits pushed, or a Jidoka stop report ending
with ## PLAN EXECUTION COMPLETE (all slices done) or a stop summary when
waiting on the developer.
Plan locations:
.planning/phases/NN-slug/*-PLAN.md.planning/quick/NNN-slug/PLAN.md(or*-PLAN.md)
Every executable unit (slice, or GSD plan wave that is one slice) must obey
Behavior | Structure, stop-safe, one observable behavior
(.cursor/rules/planning.mdc). If it does not, stop and re-plan with
slice-planning before implementing.
Git does not use the Nix prefix. All other repo tooling does:
CURSOR_DEV=true nix develop -c … unless on Cloud VM (use cloud-vm-setup
skill — no nix prefix there).
Coordinator role: You are a thin coordinator. You do not implement slices yourself (except a single interactive slice). Delegate each slice to a fresh sub-agent so context does not accumulate.
Wrap-up ownership (hard): The coordinator owns post-change-refactor,
plan update, commit, and push. Implementers must not commit and must not
run post-change-refactor themselves (nested agents routinely skip spawning a
second Task). The coordinator spawns a fresh refactor agent and must see
## REFACTOR COMPLETE (or handle ## REFACTOR JIDOKA STOP) before committing.
Resume: The PLAN file being executed is the source of truth for remaining
slices (status, learnings, adjusted later slices). Do not write
.planning/STATE.md, and do not treat it as execution or resume state.
Parallelism: Run multiple independent plans/slices in parallel (GSD waves or
Task agents) when files_modified / touch sets do not overlap and they do not
contend on the same PLAN file. Otherwise run sequentially. Each parallel
unit still gets its own coordinator-owned refactor → commit before the next
dependent unit starts.
<preflight_gate name="jidoka_stop_conditions"> Run with full autonomy but stop the line when something requires a developer's brain.
Stop and wait when:
- Value decision — multiple valid directions with different user-facing trade-offs; the plan says "TBD", "decide", "option A / B", or you discover such a fork during implementation.
- Design decision — a structural choice that affects future slices or overall architecture.
- Authentication / credentials — secrets, API keys, login flows, or permissions the agent cannot supply.
- Unexpected failure you cannot diagnose — test fails for reasons unrelated to the current change, CI breaks on something external, etc.
- Ambiguity — the slice description is unclear and guessing wrong would waste a commit.
When stopping: explain what you learned, why you stopped, and what decision the developer needs. Then wait.
Do NOT stop for:
- Routine implementation choices (naming, file placement, test structure) where existing rules and conventions give a clear answer.
- Minor refactoring needed to make the slice fit.
- Test failures caused by your own change (fix them).
Check Jidoka both before and after each slice:
- Before (coordinator, on the slice description) — safe to start autonomously? Value/design forks, ambiguity, missing credentials, Behavior/Structure grammar.
- After (implementer return + refactor return) — did work reveal something the plan did not anticipate? Stop even if the slice succeeded. </preflight_gate>
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.
- 3d ago First seen · 270 lines · 81 tokens per session scan A ffcfc2c428b6
execute-plan is a skill published in the GitHub repository nerds-odd-e/doughnut (49 stars, last pushed 3d ago), licensed MIT. It adds 81 tokens to every session and 3,144 once invoked, about $0.0004 per session on Opus 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 skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…