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/devstride/claude-plugin/pushnpx skills add devstride/claude-plugin --skill pushgit clone --depth 1 https://github.com/devstride/claude-pluginWhat 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 | $0.00018 | $0.00892 |
| Opus 5 | $0.00009 | $0.00446 |
| Sonnet 5 | $0.00004 | $0.00178 |
| Haiku 4.5 | $0.00002 | $0.00089 |
Grade A, and why
push 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.
What it actually says
Human output. Read ${CLAUDE_PLUGIN_ROOT}/skills/build-item/references/plain-language-output.md once per top-level run; composed skills reuse it. Apply it to every message.
Load .claude/ds-config.json first. Its verify, generated, protectedBranches and
commitConventions values are authoritative; absent → use stated fallbacks and say so.
Execute the following git workflow:
- Run
git add -uto stage changes to already-tracked files only (do NOT sweep in untracked files withgit add .) - Run
git statusto see what's being committed — and stage any NEW file the change requires, BY NAME (git add path/to/new-file.ts).git add -umisses new files. Never usegit add .; leave unrelated untracked files alone - Run
git diff --cached --statto show a summary of staged changes - Create a commit with a meaningful message per
commitConventions(fallback: Conventional Commits with the item tag, shaped peritemTagFormat— e.g.[I#####],[PROJ-123]— so a repo whose tracker does not use DevStride-style numbers can still express its own). The item tag applies only when the work HAS an associated item — an itemless commit (a merge, a repo chore) keeps a meaningful conventional message with no tag; never compose an item number to satisfy the format- AI attribution trailers are optional. If the active session exposes valid current-session attribution metadata, append
Co-Authored-By:plus the matching provider-specific session trailer (Claude-Session:orCodex-Session:). If that metadata is unavailable or incomplete, omit the AI trailers and continue without stopping to ask. Never invent values or reuse stale/example metadata. - The "no AI attribution" rule still applies to PR BODIES. Optional attribution belongs only on commits.
- AI attribution trailers are optional. If the active session exposes valid current-session attribution metadata, append
- Resolve the exact type-check command list (
verify.typecheckarray, in order; useverify.typecheckCombinedonly when the array is absent). A caller may pass a verification receipt. Reuse it only when every validity rule in${CLAUDE_PLUGIN_ROOT}/skills/build-item/references/verification-receipts.mdholds for the current tree and this exact list. Step 4 normally created a new commit and invalidated an older receipt, so run every command unless same-tree proof genuinely exists. With no command, ask rather than guessing. After a pass, write/refresh the receipt for downstream callers - If the type-check passes, OR the only failures are generated-file errors tolerated by
generated.toleratedTypeErrorsin files matched bygenerated.paths:- Push to remote with
git push - If push is rejected (non-fast-forward), use
git push --force-with-leaseto force push safely - On success, recap the commit, branch, push result and checks that passed or did not run.
- Push to remote with
- If TypeScript check fails with non-SDK errors:
- Do NOT push
- Report the errors and ask what to do
Do NOT create a pull request. This command stages, commits, and pushes only — PRs are opened separately when you're ready, not on every push.
Tolerated generated-file errors are defined by config: an error is tolerated only when its file
matches generated.paths AND its text matches a pattern in generated.toleratedTypeErrors.
With neither key configured, nothing is tolerated — any type error stops the push, which is the
right default: a repo that has not declared generated files has none to forgive. Where they are
configured, fix such an error by re-running generated.regenCommand, never by hand-editing the
generated file.
IMPORTANT:
- Always use force-with-lease instead of force for safety
- Only force push if the rejection is due to rewriting history on the current branch
- Never force push to main/master branches
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 · 58 lines · 18 tokens per session scan A 707f121924e1
push is a skill published in the GitHub repository devstride/claude-plugin (2 stars, last pushed 3d ago), licensed MIT. It adds 18 tokens to every session and 892 once invoked, about $0.0001 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…