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/pivyme/suiperpower/learnnpx skills add pivyme/suiperpower --skill learngit clone --depth 1 https://github.com/pivyme/suiperpowerWrote 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/skills/pivyme/suiperpower/learn)<a href="https://agentmods.dev/skills/pivyme/suiperpower/learn"><img src="https://agentmods.dev/badge/skills/pivyme/suiperpower/learn.svg" alt="Measured on agentmods" 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 | $0.00037 | $0.01506 |
| Opus 5 | $0.00018 | $0.00753 |
| Sonnet 5 | $0.00007 | $0.00301 |
| Haiku 4.5 | $0.00004 | $0.00151 |
Grade A, and why
learn 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 5d 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 — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Preamble (run first)
# Suiperpower telemetry. Routes through the CLI so projects.json + Convex stay in sync.
# Silent on failure so the skill never blocks. Opt out: SUIPERPOWER_TELEMETRY=off.
#
# AGENT NOTE: when this skill finishes, run the matching completion command:
# suiperpower track learn learn completed
# Or use "failed" / "aborted" if it ended that way.
command -v suiperpower >/dev/null 2>&1 && suiperpower track learn learn started >/dev/null 2>&1 &
true
If TEL_PROMPTED is no, before doing real work, ask the user:
Help suiperpower get better. We track which skills get used and how long they take. No code, no file paths, no PII. Change anytime in
~/.suiperpower/config.json.A) Sure, anonymous B) No thanks
Write the answer to ~/.suiperpower/config.json telemetryTier field and create ~/.suiperpower/.telemetry-prompted. Then continue.
What this skill does
Wraps a working session into a short append-only entry in .suiperpower/learnings.md, structured under the canonical headers from the phase-handoff spec. The point is a tight, honest record: what was tried, what worked, what did not, what is still open, and what was decided. Future skills (and the user themselves the next morning) read this to avoid relitigating settled decisions or repeating dead ends.
Append-only. Never deletes prior entries. Always dates new ones.
When to use it
- End of a coding session when the user wants the state captured before closing.
- After a build sprint resolved or got blocked.
- Before stepping away from a project for more than a day.
- Right after a decision was made (architecture, scope, sponsor track) so it does not leak out of context.
When NOT to use it
- The session was trivial (one-line tweak, README typo). Not everything needs a record.
- The user wants a pitch summary or a public writeup. Route to
create-pitch-deckor to a marketing skill. - The user wants to track tasks. Use a TODO file in the project, not the learnings log.
What ships with it
3 files 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.
- 5d ago First seen · 140 lines · 37 tokens per session scan A 4c49709826a2
learn is a skill published in the GitHub repository pivyme/suiperpower (12 stars, last pushed 26d ago), licensed MIT. It adds 37 tokens to every session and 1,506 once invoked, about $0.0002 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
writing-intelligence
Writing Intelligence v6 — the Sovereign Meaning Runtime. The governed craft kernel (11 passes, 27 genre packs, voiceprints, anti-slop) plus a semantic state machine — claims anchored to an exact span of an exact source version, semantic branches, proposals bound to the state they were written against, capability…
Legal Prose Polish
Use when revising draft legal work product to remove AI-sounding filler, vague risk language, unsupported certainty, and style defects while preserving citations, defined terms, quotations, privilege labels, client instructions, and attorney-review warnings.
writing-intelligence-craft
The craft library for Writing Intelligence — 27 genre packs (grant and NOFO, policy brief, journalism, medical, legal, investor memo, technical doc, book, resume, sermon, speech, newsletter, social, fiction, screenplay and more), the measurable voiceprint system, the anti-slop detector set, the positive-pattern…
anti-slop-ua
Ukrainian anti-AI-slop humanizer. Use when writing, rewriting, or reviewing Ukrainian prose (blog, landing, email, UI copy) to strip AI tells - кальки, канцелярит, бінарні контрасти - and produce живий, natural Ukrainian. Гуманізатор української прози проти ШІ-слопу.
reflect-and-improve
Use after completing significant work, recovering from errors, or at session breakpoints to review actions, synthesize learnings, and write persistent improvements to memory files. Reduces repeated mistakes and user friction over time.
zero-slop
Use when writing any prose — enforces extreme clarity by eliminating AI writing patterns, banned words, structural cliches, and filler. Produces writing that is specific, concise, and has exactly one meaning per sentence.