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/odere-pro/claude-calibration/plugin-updatenpx skills add odere-pro/claude-calibration --skill plugin-updategit clone --depth 1 https://github.com/odere-pro/claude-calibrationWrote 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/odere-pro/claude-calibration/plugin-update)<a href="https://agentmods.dev/skills/odere-pro/claude-calibration/plugin-update"><img src="https://agentmods.dev/badge/skills/odere-pro/claude-calibration/plugin-update.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 | $0.00097 | $0.00673 |
| Opus 5 | $0.00048 | $0.00336 |
| Sonnet 5 | $0.00019 | $0.00135 |
| Haiku 4.5 | $0.00010 | $0.00067 |
Grade A, and why
plugin-update 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 4d 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.
What it actually says
plugin-update
After /docs-update refreshes docs/, this skill propagates any changed facts into the per-feature
bundles so the rubric the plugin grades against and the rubric it ships stay identical.
Scope
With no argument, process all nine calibrate-<feature> bundles. With an argument
(/plugin-update skills), process just that feature.
Steps
- Pair each bundle with its source:
skills/calibrate-<feature>/reference.md↔docs/features/<feature>.md(eachreference.mdalready cites its source-of-truth page). - Diff the encoded facts between the docs page and the bundle's
reference.md,templates/*, and the constants inscripts/lint.sh(e.g.NAME_MAX,DESC_MAX,BODY_MAX, threshold values embedded in signatures likeover-200). - Respect the signature contract. If a limit moved (e.g. body cap 500 → 600), the signature name
carrying the threshold (
skill:body-over-500) changes too — and that means the four-places update from../../rules/CLAUDE.md:rules/signatures.md, the bundle'sreference.md, itsscripts/lint.shemit, andrules/dispatch.md. Never rename a signature silently — call it out. - Propose edits, grouped by bundle, and wait for approval before writing.
- Verify after each bundle: re-run its
scripts/lint.shon its own templates/examples to confirm it still emits cleanly. - Propose a version bump. Summarise what changed and recommend a
plugin.jsonversionbump (patch for limit tweaks, minor for a new signature/check). Leave the actual bump + changelog fragment to the user — this skill does not editplugin.jsonor tag a release.
Boundaries
- Edits bundle files under
skills/calibrate-*/andrules/only; neverplugin.json, never a tag. - The plugin-dev write-guard (
.claude/hooks/plugin-dev-guard.sh) still applies: arules/file must keep itspaths:frontmatter.
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.
- 4d ago First seen · 48 lines · 97 tokens per session scan A ba81bc28b16c
plugin-update is a skill published in the GitHub repository odere-pro/claude-calibration (1 stars, last pushed 1mo ago), licensed MIT. It adds 97 tokens to every session and 673 once invoked, about $0.0005 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
deploy-setup
Wire up a freshly scaffolded Starter Series project for its first release — detect the template, register required GitHub secrets, set up OIDC trusted publishing where applicable, and trigger the first CD run. Pairs with the create skill (you scaffold first, then run this).
create
Scaffold a new project from the Starter Series templates (MCP server, Discord/Telegram bot, VS Code / browser extension, Electron, React Native, Cloudflare Pages, npm package, Docker deploy).
routing
Select which repo specialist agents to involve for a given ticket. Uses keyword scoring with synonym expansion, acronym normalization, and a confidence signal passed to each specialist.
workspace-knowledge
Answer questions about the whole workspace — which repo owns or handles something, where a feature/endpoint/event lives, or how concerns span repos. Reads registry.json as the ownership index, routes the question to the owning repo, and hands off to that repo's specialist for detail. Use for cross-repo and 'which…
fastapi-conventions
FastAPI routing, dependency injection, and async patterns. Use when building or reviewing FastAPI backends.
react-patterns
Idiomatic React — hooks, composition, and clear state boundaries. Use when writing or reviewing React components.