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/kreek/consult/commitnpx skills add kreek/consult --skill commitgit clone --depth 1 https://github.com/kreek/consultWhat 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 | $0.00013 | $0.00987 |
| Opus 5 | $0.00006 | $0.00494 |
| Sonnet 5 | $0.00003 | $0.00197 |
| Haiku 4.5 | $0.00001 | $0.00099 |
Grade A, and why
commit 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.
How it starts
The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Commit
Iron Law
COMMIT ONLY THE REVIEWED SLICE. STAGE FILES BY NAME.
Commit packaging is common and should stay lightweight. The host owns the git
invocation mechanics; this skill sets the packaging standards. Load
git-workflow only when branches, conflicts, rebases, history edits,
recovery, or GitHub CLI use are part of the task.
When to Use
- The user asks to commit, stage, split, or package current changes.
- You need to group dirty files into one or more logical commits.
- You need a right-sized commit subject or body for approved work.
When NOT to Use
- Creating or switching branches, resolving conflicts, rebasing, bisecting,
recovering history, deleting branches, or force-pushing. Use
git-workflow. - Reviewing correctness of the diff. Use
code-reviewbefore packaging non-trivial implementation work. - Preparing release versions, changelogs, tags, or publishing. Use
releaseonly when requested or approved.
Core Ideas
- A branch is not the commit boundary. Even on one branch, group changes by behavior so reviewers can understand what changed and maintainers can revert one behavior without dragging unrelated work with it.
- Logical commits are reader-facing history. A commit should explain one completed behavior, cleanup, or documentation change; nearby work can wait for a second commit when it would need a different rollback decision.
Workflow
- Check the tree state and stop on anything unexpected: wrong branch, unresolved merge state, or files you do not recognize.
- Separate unrelated work by behavior, even when all changes belong to the
same branch. Stage only named files or approved pathspecs for the reviewed
slice; never use
git add .in a messy tree. Confirm the staged diff matches the slice before committing. - Confirm relevant proof is current. If a broad suite is noisy for unrelated reasons, name the targeted proof and report the broader drift separately.
- Write a commit message that completes "When applied, this commit will ...".
Use a concise subject. Add a body only when the change needs context, up to
2-3 short paragraphs. Attribution trailers are host configuration, not
message content: when a host adds trailers such as
Co-Authored-Byby default, point the user at the host setting that disables them (for exampleincludeCoAuthoredBy: falsein Claude Code settings) instead of overriding the host's guidance from inside the message. - Commit without skipping hooks.
What ships with it
1 file 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.
- yesterday First seen · 91 lines · 13 tokens per session scan A 7c0c8f125d39
commit is a skill published in the GitHub repository kreek/consult (1 stars, last pushed 6d ago), licensed MIT. It adds 13 tokens to every session and 987 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
peekaboo
Capture and automate macOS UI with the Peekaboo CLI.
oracle
Best practices for using the oracle CLI (prompt + file bundling, engines, sessions, and file attachment patterns).
argent-tv-interact
Control and inspect TV apps via argent — Apple TV (tvOS), Android TV (leanback), and Amazon Fire TV (Vega). Boot the target, read focus, navigate with the D-pad remote, type, screenshot, and on Vega debug the JS runtime (evaluate, console logs, network inspector). Use when a task targets a TV (runtimeKind "tv", or…
review
5-pass structured code review — correctness, security, performance, readability, consistency.
live-preview
Mid-build visual verification loop. Takes screenshots of components during construction, not just after. Catches visual regressions and invisible features before they compound. Requires Playwright or similar screenshot tool.
wiki
Markdown-first knowledge base where the LLM acts as librarian. Ingests raw sources, compiles and interlinks topic files, self-maintains an index. No vector DB or embeddings required -- uses LLM-native navigation over structured markdown up to 400K words.