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/emaballarin/ccplugins/plannpx skills add emaballarin/ccplugins --skill plangit clone --depth 1 https://github.com/emaballarin/ccpluginsWhat 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 | $0.00199 | $0.01365 |
| Opus 5 | $0.00100 | $0.00682 |
| Sonnet 5 | $0.00040 | $0.00273 |
| Haiku 4.5 | $0.00020 | $0.00136 |
Grade A, and why
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 2d 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 — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/tml:plan — decide where to sit, and how long to train
Two decisions, both the operator's: where on the frontier this project sits, and how long each run gets. This skill frames them, prices the options, and asks. It does not choose.
First action, always
ls -la ./.tml/ 2>/dev/null
Read findings.md if present. If it is absent, say so and either run
/tml:audit first or proceed from what the operator states — but never invent
findings to plan against.
Hard rules
- The operating point is not the agent's to choose. Present the options with honest upsides and downsides and ask. Silence is not consent.
- A quality floor is mandatory before any
spendingchange is ordered. No floor, no spending. - Non-negotiables are vetoes. A hard constraint touched by an
unknownorspendingexposure stops that change regardless of effect size. - Every ordered change carries a revert trigger — the observation that means undo it — written before it is made, not after it fails.
- Writes only
./.tml/frontier.md. No project code.
Procedure
1. Establish the regime
references/regime.md §1 and §3: concurrent trials, local or remote execution.
This is not optional and it is not inferable — a plan calling for 60-trial
studies on a machine that runs two at a time is not a plan. Record what was
sacrificed if the regime is low (study-design.md §4).
2. Fix the step budget
references/step-budget.md. Determine compute-bound or not, then:
- Not compute-bound — pick
max_train_steps(§2, including the constant-LR sweep procedure and its self-deception failure mode) and fix it across all trials. Never tune it inside a study. - Compute-bound — plan rounds of increasing per-trial length (§3), and carry the transfer ladder (§4) so it is clear which round-1 conclusions are expected to survive: warmup and initialisation very likely, the decay schedule unlikely.
3. Name the non-negotiables
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.
- 2d ago First seen · 113 lines · 199 tokens per session scan A 53278051ac4f
plan is a skill published in the GitHub repository emaballarin/ccplugins (3 stars, last pushed 26d ago), licensed MIT. It adds 199 tokens to every session and 1,365 once invoked, about $0.0010 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…