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 dwarvesf/dwarves-kit --skill doc-driftgit clone --depth 1 https://github.com/dwarvesf/dwarves-kitWrote 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/dwarvesf/dwarves-kit/doc-drift)<a href="https://agentmods.dev/skills/dwarvesf/dwarves-kit/doc-drift"><img src="https://agentmods.dev/badge/skills/dwarvesf/dwarves-kit/doc-drift/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/dwarvesf/dwarves-kit/doc-drift"><img src="https://agentmods.dev/badge/skills/dwarvesf/dwarves-kit/doc-drift.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 | $0.00163 | $0.01466 |
| Opus 5 | $0.00081 | $0.00733 |
| Sonnet 5 | $0.00033 | $0.00293 |
| Haiku 4.5 | $0.00016 | $0.00147 |
Grade A, and why
doc-drift 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 — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Doc drift
Overview
Audit every living doc against the live repo and ship the fixes as a PR. This is the doc-drift
instance of docs/patterns/audit-loop.md: enumerate, verdict with evidence, apply on a branch,
gate through a PR the operator approves. /kit:docs syncs docs against ONE diff during a build
cycle. This skill audits the WHOLE estate on a cadence, and catches what diff-scoped sync
structurally misses: drift from merged-but-undocumented work, inventory rot, dead pointers,
stale counts. PR #306 (hand-fixing the README inventories) is the proven need.
The four slots (per the audit-loop pattern)
| Slot | This instance |
|---|---|
| Item set | living docs only: git ls-files README.md MANUAL.md AGENTS.md CLAUDE.md CONTRIBUTING.md ':(glob)docs/*.md' ':(glob)docs/patterns/*.md' (the :(glob) magic is required: a bare docs/*.md pathspec matches recursively and pulls in every dated record) |
| Contract | every checkable claim matches the live repo: referenced paths exist, inventory tables match the live dirs, counts match derived counts, named verbs and flags exist, cross-doc pointers resolve |
| Evidence class | Tier 1: the repo itself via grep/ls/diff. Tier 2: a model reads the doc's described behavior against the actual command or script |
| Apply mechanics | FIX in place on an isolated branch; REMOVE only with a named successor; PR body lists every verdict; UNSURE never auto-resolved |
Dated records (docs/specs, docs/research, docs/retro, docs/handoff, docs/decisions, docs/absorption, and docs/CHANGELOG.md's per-release entries) are OUT of the item set. They describe their moment. Flagging a 2026-05 spec for describing 2026-05 reality is noise, not a finding.
Process
-
Branch first. All edits ride an isolated branch. Auditing on master is the failure this skill exists to prevent.
-
Enumerate. Run the item-set command above. Write the list down before judging anything. The list is the queue; a resumed or scheduled run picks up where the last one stopped.
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 · 92 lines · 163 tokens per session scan A e6259e594f4b
doc-drift is a skill published in the GitHub repository dwarvesf/dwarves-kit (11 stars, last pushed today), licensed MIT. It adds 163 tokens to every session and 1,466 once invoked, about $0.0008 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 2026-09-04.
Other skills, from other repositories
self-audit
Self-audit this skills repo against CLAUDE.md invariants, the artifact contract, and README/CLAUDE.md/docs sync via three parallel read-only subagents. Local meta-skill — independent of the /task: pipeline.
refactorlib-cc
Audit a codebase for handcrafted code that duplicates functionality already available in the project's dependencies. Reads package.json, launches parallel exploration agents, verifies replacement feasibility, and produces a structured refactor plan. Audit only -- does not execute changes.
triage
Verify an open GitHub issue or PR against the current code and decide its fate: implement, keep, or close with a factual comment. Use when the user asks to triage, verify, check, handle, or clean up an issue, a PR, or the tracker.
pr-feedback-ingest
Turn PR feedback (Greptile, CI, bots, humans) into a structured traceable backlog aligned with TASKSTATE.md, DECISIONS.md, and IMPLEMENTATIONPLAN.md so the next execution step can be a narrow implement-approved-slice or a small planning touch without losing alignment. Corrective scope only. Use when a PR is open or…
repo-consistency-sweep
Proactive defect-class detection that handles the lower-value half of code review (per Bacchelli and Bird 2013) so human reviewers stay focused on design, intent, and knowledge transfer. Catches convention drift, ordering bugs, type-safety gaps, security and multi-tenant invariants (CWE-grounded), and operability…
review-depth
Decide how much review a change has earned, before reviewing it. Selects quick, standard, or deep from stated signals (reviewable changed lines, directory spread, trust-boundary risk flags, blast radius, documentation-only scope), reports the signal table and the rule that fired, then runs house-review at the selected…