fable-audit

fable-audit is a skill for Claude Code, Codex from deepklarity/harness-kit. It costs 86 tokens per session (593 once invoked), scanned A, original, MIT.

A roadmap auditor for Fable, checking documented development progress against the project board and completed work. A roadmap is a plan for what a project intends to build and when.

In plain words
What is it for?
Use it when a roadmap wave ends or when you need a brief progress audit, updated resume state, and reprioritized backlog.
Why use it?
It helps separate verified progress from unsupported claims and exposes differences between project records, documentation, evidence, and scoring rules.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/deepklarity/harness-kit/fable-audit
Any agent
npx skills add deepklarity/harness-kit --skill fable-audit
Clone the repo
git clone --depth 1 https://github.com/deepklarity/harness-kit

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for fable-audit

README.md
[![agentmods](https://agentmods.dev/badge/skills/deepklarity/harness-kit/fable-audit.svg)](https://agentmods.dev/skills/deepklarity/harness-kit/fable-audit)
Your own site
<a href="https://agentmods.dev/skills/deepklarity/harness-kit/fable-audit"><img src="https://agentmods.dev/badge/skills/deepklarity/harness-kit/fable-audit.svg" alt="Measured on agentmods" height="20"></a>
Per session 86 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 593 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00086 $0.00593
Opus 5 $0.00043 $0.00296
Sonnet 5 $0.00017 $0.00119
Haiku 4.5 $0.00009 $0.00059

Measured 5d ago against content hash 40f89ddb6532, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

fable-audit 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.

.claude/skills/fable-audit/SKILL.md · 38 lines

What it actually says

Fable roadmap auditor

Trust nothing; verify everything. Lean by user directive: an audit is one short report plus two file updates — not a documentation pass.

  1. Verify claims with proof. For each task DONE since the last audit, run its verify step or inspect the merged artifact. Delegate command runs and file checks to parallel haiku subagents; keep judgment in the main conversation. The board is truth (board_overview.py / spec_trace.py / task_inspect.py --brief from taskit/taskit-backend/); doc-vs-board divergence is a finding. Downgrade anything unprovable.
  2. Recompute what has data: run the suites; count autonomy/unsticks from board history (hand-counted until the metrics script lands — say so). Report gaps as gaps.
  3. Grade against SCORECARD.md — every score move names the anchor (0/5/10) its evidence supports; recompute the overall (plain average). Scores live only in the fable_roadmap.md table — a score anywhere else is a finding. Then judge the ladder against fable_roadmap.md exit criteria; check the ratchet rules there. Violations are findings.
  4. Write audits/YYYY-MM-DD*.md: verdict line (on track / drifting / blocked + why), verified-done table with proof, demotions, metrics snapshot, top-3 risks with mitigations, ladder call. Readable in 30 seconds from the top.
  5. Update the living docs: RESUME "Current state" + BACKLOG re-rank (the top IS the next wave draft) + the ladder Status column if a level changed.
  6. Artifact only if the user asks — redeploy to the URL in RESUME's durable facts; never mint a new one.
  7. --dispatch only: convert the backlog top into a wave spec under bootstrap/ and load it (pre-dispatch checklist in the loader templates).

End with: verdict line, ladder level, top 3 next actions — plain prose.

Changes

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.

  1. 5d ago First seen · 38 lines · 86 tokens per session scan A 40f89ddb6532

Subscribe to this mod's changes

fable-audit is a skill published in the GitHub repository deepklarity/harness-kit (95 stars, last pushed 1mo ago), licensed MIT. It adds 86 tokens to every session and 593 once invoked, about $0.0004 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 2026-08-30.