audit
Skill0 starsChanged 3 months agoMITView on GitHub
Use when you want to check whether a running or finished spiral stayed risk-driven or became an incremental build wearing spiral clothing. Audits a spiral against the risk-driven invariants the binary cannot enforce: scores that actually rank, gate evidence that actually retires risks, exposure that actually drains, a stop that gets considered. Verdicts cite the ledger.
194 other projects use a version of this.
Adopted by
Most recently taken up by
- Laucked-Security/claude-oswetheir own version, first seen Nov 8, 2026
- sanmidable/skillstheir own version, first seen Nov 8, 2026
- alegauss/claude-traytheir own version, first seen Nov 8, 2026
- ansvisor/ansvisortheir own version, first seen Nov 8, 2026
- ai-search-guru/getcito-worlds-first-open-source-aio-aeo-or-geo-tooltheir own version, first seen Nov 8, 2026
- MammadovEshgin/clean-code-skilltheir own version, first seen Nov 8, 2026
Used with
Projects with it also use delight, teach-impeccable and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
HK-hubSkilldelight
Add moments of joy, personality, and unexpected touches that make interfaces memorable and enjoyable to use. Elevates functional to delightful. Use when the user asks to add polish, personality, animations, micro-interactions, delight, or make an interface feel fun or memorable.
Also in 32 projects that use this
HK-hubSkillteach-impeccable
One-time setup that gathers design context for your project and saves it to your AI config file. Run once to establish persistent design guidelines.
Also in 19 projects that use this
zwy123zwySkillnormalize
将用户随意输入的提示词规范化为结构清晰、便于大模型执行的标准化提示词。适用于需要整理、统一提示词格式时调用。
Also in 19 projects that use this
SkillPilotCHSkilllevel-up
Eine Automatisierung pro Woche finden, scopen, shippen. Output landet als proposed-Eintrag in cadence.md mit konkreten Implementations-Schritten.
Also in 12 projects that use this
ww-w-aiSkillpm-discovery
PM Agent Team - Automated product discovery, strategy, and PRD generation. Runs PM analysis workflow to produce comprehensive PRD before PDCA Plan phase.
Also in 3 projects that use this
- +7 more from HK-hub
Other versions194 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 194
- openai/plugins★ 7.8ksame name
- ccch1mneyyy/dsh-TUI★ 7.2ksame name
- macro-inc/macro★ 6ksame name
- yournextstore/yournextstore★ 5.5ksame name
- gadievron/raptor★ 4ksame name
- alchemy-run/alchemy★ 1.9ksame name
- alchemy-run/alchemy-effect★ 1.9ksame name
- mattmireles/gemma-tuner-multimodal★ 1.5ksame name
- jsmastery-pro/skills★ 1.4ksame name
- nateherkai/AIS-OS★ 1.4ksame name
184 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 1 is distinctive, and 1 is very specific to this project.
0
Common
Most of the 176 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 23
Use it mid-spiral as a health check, or after stop/drain as a post-mortem.
- Distinctive: line 38
Read the whole ledger first, then walk the table.
“Common” is measured: kept by at least 6 in 10 of the 176 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
murex - Spiral Audit
Description
You audit a spiral run for the discipline the binary cannot enforce. murex already guarantees the mechanics: every cycle targets the top-exposure open risk, a pending cycle must close through a gate, risks are never deleted. What it cannot check is the human side - whether the numbers, evidence, and decisions fed into it were honest. That is this skill.
The failure mode it hunts: an incremental build wearing spiral clothing.
Cycles that ship features and call them spikes, gates rubber-stamped
continue, a register scored once at the start and never touched again. The
spiral model is risk-driven - what each cycle does is chosen by the largest
remaining unknown, not by "what to build next". A spiral that stops being
risk-driven is just a slower waterfall, and this audit says so.
Use it mid-spiral as a health check, or after stop/drain as a post-mortem.
Prerequisites
A ledger: .murex/spiral.json in the target repo. Read the file directly -
it is plain JSON and the audit needs every field - and cross-check the
derived numbers with the binary:
murex --root . status # radius, remaining_exposure, history
murex --root . risk list # open (ranked) and closed risksThe checks
Read the whole ledger first, then walk the table. Every verdict must cite ledger fields - risk ids, cycle numbers, exposures, evidence strings. A finding without a citation is not a finding.
| # | Check | Red flag in the ledger |
| 1 | Scores rank | probability/impact clustered on one value across risks - the exposure ranking carries no information, so "top risk" was chosen by accident |
| 2 | Evidence retires | risks closed resolved whose evidence has no number, artifact, or path in it; gate outcome empty |
| 3 | Exposure drains | open exposure flat or rising across two closed cycles while cumulative_cost grows - spikes are not producing evidence, or features are being built under a spike's name |
| 4 | Cost is real | closed cycles with cost: 0 - the radius is fiction and the commitment review compared against nothing |
| 5 | Unknowns land in the register | several cycles in, yet no risk has cycle_opened > 0 - implementation discovered nothing? More likely discoveries bypassed the ledger |
| 6 | Accepted means decided | accepted risks with empty evidence - shipped past knowingly, but the reason is not auditable |
| 7 | Stop gets considered | remaining exposure still high after repeated cycles or a pivot, every gate says continue - "do not build" is a legitimate outcome nobody weighed |
| 8 | The spiral earns its cost | every open exposure marginal - no real unknowns left; the correct move is to leave the spiral and just build |
For check 3, reconstruct the exposure timeline from the risks themselves:
a risk counts toward cycle N's open exposure when cycle_opened <= N and
(cycle_closed is null or > N).
Verdict
One line per check - PASS or FINDING, citation attached:
1 scores-rank FINDING 6 of 7 risks scored 0.5/0.5; only R3 differs
2 evidence-retires PASS R1 "380MB RSS", R4 "auth handshake trace"
3 exposure-drains FINDING cycles 3-4: exposure 1.12 -> 1.12, cost +2.5
...Then a stance, not a summary: the single most damaging finding, and the
smallest fix for it. Map fixes to the existing surface only - re-score
(murex risk add was honest, the numbers were not), record the missing
evidence (murex risk close --evidence), register the bypassed unknowns
(murex risk add), take the gate seriously next murex commit, or leave
the spiral (murex stop --reason).
Boundaries
- Never edit
.murex/spiral.jsonby hand. The audit reads the ledger; every correction goes through the binary or the human. - The audit judges the run, not the objective. Whether the goal is worth pursuing belongs to the commitment review; whether the review was honest belongs here.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

