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 skills add shahidshabbir-se/my-pi-setup --skill self-improvegit clone --depth 1 https://github.com/shahidshabbir-se/my-pi-setupWrote 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/shahidshabbir-se/my-pi-setup/self-improve)<a href="https://agentmods.dev/skills/shahidshabbir-se/my-pi-setup/self-improve"><img src="https://agentmods.dev/badge/skills/shahidshabbir-se/my-pi-setup/self-improve/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/skills/shahidshabbir-se/my-pi-setup/self-improve"><img src="https://agentmods.dev/badge/skills/shahidshabbir-se/my-pi-setup/self-improve.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 | $0.00120 | $0.00609 |
| Opus 5.5 | $0.00048 | $0.00244 |
| Sonnet 5.5 | $0.00024 | $0.00122 |
| Haiku 4.5 | $0.00012 | $0.00061 |
Grade A, and why
self-improve 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 19d 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 — 35 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Self-Improve
Turn a concrete agent failure into a versioned rule. $ARGUMENTS may name the mistake and the desired rule; if it doesn't, ask before writing anything.
When NOT to use
- One-off typos or a single bad guess with no reuse value — just fix it, don't write a rule.
- The failure is really "the product/spec was ambiguous" and the user hasn't confirmed a preference yet — ask first instead of encoding a guess as a permanent rule.
Steps
-
Pin the failure — One sentence: what the agent did wrong, and the correct behavior instead. If it doesn't clear the bar above, stop here.
-
Pick the home (first match wins):
- Applies every session / every prompt in this repo → one bullet in the root
AGENTS.md, under the tightest existing heading (or a new short heading only if nothing fits). - Applies to one app or package only → a nested
AGENTS.mdinside that package (create it only if the rule is likely to recur there). - Describes a multi-step workflow rather than a single rule → extend or add a skill under
.agents/skills/instead of bloatingAGENTS.md.
- Applies every session / every prompt in this repo → one bullet in the root
-
Write the rule — Positive, short, checkable. Match the existing tone of
AGENTS.md. No essays. Prefer "do X" over a long list of "don't do Y." If the rule is really "look here for the answer," point at the path rather than restating the content. -
Dedupe — Search
AGENTS.mdand nearby agent docs for the same idea already written down. Edit in place rather than adding a duplicate. -
Keep it lean — If the root
AGENTS.mdis drifting long, move branch-specific detail into a skill or a nested file and leave a one-line pointer from the root. -
Report — State the file changed, the exact rule text added or edited, and why it belongs there.
Completion
Done when the durable rule is committed to the right file in the working tree, and any redundant wording elsewhere has been removed. Do not create a rule for a preference the user hasn't confirmed when the failure is product-ambiguous — ask first, per the guardrail above.
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.
- 19d ago First seen · 35 lines · 120 tokens per session scan A 625b5a8a2c33
self-improve is a skill published in the GitHub repository shahidshabbir-se/my-pi-setup (2 stars, last pushed 2d ago), licensed MIT. It adds 120 tokens to every session and 609 once invoked, about $0.0005 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 2026-09-27.
Other skills, from other repositories
repo-hygiene
Use when the scheduled repo-hygiene workflow runs from GitHub Actions (or an operator dry-run) to scan the repository for small, certain docs/test/code hygiene issues and fix them as one batched branch.
browser-use
Control the user's Chrome through the existing persistent Node REPL and the Qwen Browser SDK.
test-failure-explorer
Import and query an existing Vitest 3.2 JSON report to explore test failures, file-level errors and original diagnostic evidence.
google-agents-cli-scaffold
This skill should be used when the user wants to "create an agent project", "start a new ADK project", "build me a new agent", "add CI/CD to my project", "add deployment", "enhance my project", or "upgrade my project". Part of the agents-cli skills suite. Covers agents-cli scaffold create, scaffold enhance, and…
find-simplifications
Use for a periodic repo-wide sweep of qwen-code for accumulated excess surface — dead components and files, orphaned locale keys, exports nothing consumes, added-then-removed scaffolding — filing candidates on a tracking issue and landing only what a maintainer has said yes to. Repo-wide and evidence-first; every…
ccc
This skill should be used when code search is needed (whether explicitly requested or as part of completing a task), when indexing the codebase after changes, or when the user asks about ccc, cocoindex-code, or the codebase index. Trigger phrases include 'search the codebase', 'find code related to', 'update the…