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/vrennat/developersdevelopers/plan-hunternpx skills add vrennat/developersDevelopers --skill plan-huntergit clone --depth 1 https://github.com/vrennat/developersDevelopersWhat 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 | $0.00127 | $0.00857 |
| Opus 5 | $0.00063 | $0.00428 |
| Sonnet 5 | $0.00025 | $0.00171 |
| Haiku 4.5 | $0.00013 | $0.00086 |
Grade A, and why
plan-hunter 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 yesterday.
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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Plan Hunter
One polished implementation plan via a four-phase tournament: Scope → Draft (×4 parallel) → Judge (×4 parallel) → Synthesize. About 10 subagents and 4 of your turns end-to-end.
When it pays off
Run when there's real planning weight: multi-week scope, sequence/scope/risk tradeoffs, or explicit user request. Skip for bug triage, tactical "how do I X", single-component choices, or anything answerable in two paragraphs — burning 10 subagents on a small question is the worse failure mode.
The phases
- Scope (1 subagent) — normalize the idea into a JSON CONTEXT block (
normalized_idea,goals,constraints,assumptions,open_questions). Every downstream agent consumes it verbatim. - Draft (4 subagents, parallel — launch in a single turn) — each commits to one lens:
- A. MVP-first — smallest shippable thing that delivers the core promise.
- B. Risk-first — sequence so the riskiest assumptions get spiked or proven early.
- C. Dependency-first — build the dependency graph; surface critical path and parallelizable tracks.
- D. User-first — work backward from the user journey; what makes the product feel real at each milestone.
- Judge (4 subagents, parallel — identical prompts) — each scores all four drafts on completeness, practicality, risk_awareness, sequencing (1–10 each). Variance across judges is the point — averaging cuts single-judge noise.
- Aggregate (no subagent) — mean per axis per plan, total mean per plan. Highest = winner; the rest are runner-ups. Collect rationales, union risks, union gaps.
- Synthesize (1 subagent) — polish the winner, graft clearly-better moves from runner-ups (note "(borrowed from {lens} lens)"), prepend assumptions + open questions. No averaging or compromise — pick the better move and justify.
Invocation
- Slash command:
/plan-hunter <idea>—$ARGUMENTSis the idea. - Auto-trigger: on substantive planning asks. The idea is the user's most recent planning message; don't ask them to repeat it.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- yesterday First seen · 47 lines · 0 tokens per session scan A c92ace107611
plan-hunter is a skill published in the GitHub repository vrennat/developersDevelopers (2 stars, last pushed 1mo ago), licensed MIT. It adds 127 tokens to every session and 857 once invoked, about $0.0006 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
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…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…