Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add melodic-software/claude-code-plugins/plugin install docs-hygieneWrote 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/melodic-software/claude-code-plugins/audit-derivability)<a href="https://agentmods.dev/skills/melodic-software/claude-code-plugins/audit-derivability"><img src="https://agentmods.dev/badge/skills/melodic-software/claude-code-plugins/audit-derivability/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/melodic-software/claude-code-plugins/audit-derivability"><img src="https://agentmods.dev/badge/skills/melodic-software/claude-code-plugins/audit-derivability.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 | $0.00124 | $0.05834 |
| Opus 5.5 | $0.00050 | $0.02334 |
| Sonnet 5.5 | $0.00025 | $0.01167 |
| Haiku 4.5 | $0.00012 | $0.00583 |
Grade A, and why
audit-derivability scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 192 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository context. Gather first
Collect these with individual Bash calls, one command per call, never combined into a single invocation:
- Current branch,
git branch --show-current
Treat a failure (not a repository, git unavailable) as an unknown value and carry on. Keep these as
separate body Bash calls rather than pre-compute lines: the harness runs a skill's whole pre-compute
block as one shell invocation, and a worktree-isolated session refuses a compound command that
contains git. The dated record for that composition claim is the source-control plugin's
worktree/reference/gather-block.md,
"The pre-compute block runs as one shell invocation".
Pre-computed context
Uncommitted .md files (first 20; empty = none matched or the probe returned nothing): !"${CLAUDE_SKILL_DIR}/scripts/uncommitted-md.sh" 20 2>/dev/null || echo "(status unavailable)"
Every command in this line matches an allowed-tools grant, so a dontAsk lane can run this skill.
- Pointer: when changing this line or its grants, fetch https://code.claude.com/docs/en/skills#permission-checks-on-injected-commands live; the probe that found a zero-turn run under dontAsk is in https://github.com/melodic-software/claude-code-plugins/pull/6732. As of: 2026-10-11. Recheck trigger: a Claude Code release note that changes how pre-compute commands are permission-checked.
Purpose
Docs carry a standing tax: every tracked document must be kept true as the code it describes moves, and every low-signal document a reader (human or agent) wades through costs attention. A document earns that tax only when it holds something a reader could NOT cheaply reconstruct for themselves. This skill audits one axis the siblings do not: document-level worth. Should this whole document exist at all?
The test is derivability: could a fresh agent reach this document's conclusions by natively exploring the repository, reading the code, config, metadata, and structure, without being handed the document? A document that only restates what the code already says is a derivation cache at best and dead weight at worst; a document that records why a decision was made, a constraint that is not visible in any single file, or a fact that lives outside the repo owns something exploration can never recover.
What ships with it
7 files 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.
- context/rubric.md 16 KB
- evals/evals.json 20 KB
- evals/fixtures/decision-rationale.md 1.0 KB
- evals/fixtures/derivable-with-source/runtime-settings.md 241 B
- evals/fixtures/derivable-with-source/settings.json 210 B
- scripts/uncommitted-md.sh 827 B runs code
- scripts/uncommitted-md.test.sh 3.3 KB runs code
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 Changed 949ca3b091af
- 6d ago Changed · +5 lines 0dd7964665c2
- 12d ago Changed 2682b81dd647
- 17d ago Changed · +2 lines 11cb809835fd
- 20d ago Changed · +6 lines b412adcb2385
- 21d ago Changed · -73 tokens per session 29215b93abcf
- 1mo ago Changed · -5 tokens per session 13c1759e11cc
- 1mo ago Changed · +2 lines a7e5c8a65fea
- 1mo ago Changed · +11 lines adceaf36d762
- 1mo ago First seen · 166 lines · 202 tokens per session scan A cee93447e1d9
audit-derivability is a skill published in the GitHub repository melodic-software/claude-code-plugins (26 stars, last pushed today), licensed MIT. It adds 124 tokens to every session and 5,834 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 2026-09-03.
Other skills, from other repositories
parallel-orchestrator
Manage parallel Claude Code workstreams using git worktrees. Use when: splitting large tasks across multiple workers, coordinating parallel development, monitoring worker progress, integrating completed work, analyzing work item documents (code reviews, issue lists). Triggers: parallel, orchestrator, worktrees…
parallel-worker
Execute focused implementation tasks in a parallel workflow. Use when: working on assigned files in a worktree, making checkpoint commits, signaling dependencies or blockers, completing orchestrator-assigned tasks. Triggers: worker, checkpoint, worktree, assigned scope, commit prefix, parallel task.
optimize-local-search
For NP-hard optimization: TSP, scheduling, assignment problems. Uses greedy construction + local improvement (2-opt, hill climbing).
solve-constraint-puzzle
For constraint satisfaction: Sudoku, scheduling, N-queens, logic puzzles, SAT-like problems, assignment problems. Uses propagate-then-search pattern.
test-with-examples
For example-driven development: test cases as specifications, input/output pairs, documentation through examples.
return-none-for-failure
For graceful failure: return None or False instead of exceptions, let caller decide how to handle failure.