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 OutlineDriven/outline-driven-development --skill ax-auditgit clone --depth 1 https://github.com/OutlineDriven/outline-driven-developmentWrote 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/outlinedriven/outline-driven-development/ax-audit)<a href="https://agentmods.dev/skills/outlinedriven/outline-driven-development/ax-audit"><img src="https://agentmods.dev/badge/skills/outlinedriven/outline-driven-development/ax-audit/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/outlinedriven/outline-driven-development/ax-audit"><img src="https://agentmods.dev/badge/skills/outlinedriven/outline-driven-development/ax-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 37 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00039 | $0.00951 |
| Opus 5.5 | $0.00016 | $0.00380 |
| Sonnet 5.5 | $0.00008 | $0.00190 |
| Haiku 4.5 | $0.00004 | $0.00095 |
Grade A, and why
ax-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 1mo 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- ax-audit — 100% identical, 5 lines differ
How it starts
The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AX audit
Contract
| Field | Bound contract |
|---|---|
| Trigger | Human asks for an agentic experience audit, an AX review, whether something is agent-native, a critique of an AI feature, or whether it earns user trust |
| Authority | Read-only. No file, VCS, credential, paid, published, deployed, or remote mutation. |
| Side effect | Chat output only: a read-only severity-tiered report with a ship-readiness verdict and an AX Relationship Summary |
| Done | Delivered findings with locators (file:line for source, logical bounds for described behavior), an AX Relationship Summary, and a PASS/FAIL/INCOMPLETE verdict |
Inputs
- Target agent system or AI feature under audit: source path, repository, or described behavior. Required.
- Scope boundary: which files, modules, or interaction surfaces are in scope. Required.
- Specific concern or focus area the human wants prioritized. Optional.
Procedure
-
Read the target source or described behavior within the stated scope. Do not mutate anything. Done when: the target is read within the scope boundary.
-
Evaluate the target against two audit axes, each with four rule families. Record every finding with a locator:
file:linefor source-backed targets, or logical bounds (described behavior, interaction surface, or module name) for behavior-only targets.Architecture axis (rules-arch):
- Communication: approval gate present where the agent should act autonomously; completion signal missing or ambiguous; progress not visible to the user during long work.
- Context: no checkpoint or resume across sessions; context injection absent or brittle; context starvation where the agent lacks information it needs to act.
- Granularity: static API mapping where tools mirror raw endpoints instead of workflow-shaped actions; tools too coarse or too fine for the task shape.
- Parity: CRUD coverage incomplete; tool or UI parity gaps where an action exists in one surface but not the other; orphaned UI actions with no backing tool.
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.
- 1mo ago Changed · +1 lines 09af55a10f5f
- 1mo ago Changed · -34 tokens per session 9e98e9b147f0
- 1mo ago First seen · 58 lines · 73 tokens per session scan A a7216b30947f
ax-audit is a skill published in the GitHub repository OutlineDriven/outline-driven-development (54 stars, last pushed 28d ago), licensed Apache-2.0. It adds 39 tokens to every session and 951 once invoked, about $0.0002 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
create-pull-request
Use when asked to create or update a PR, revise its description, or link issue references to its body. Not for multi-PR stacks: use gate-and-merge. Not for releases: use git-workflow-and-versioning.
gate-and-merge
Use when landing a queue of open PRs: gate each PR, sweep its review feedback to root cause, then merge, repair, hold, or close it. Human-only.
doubt-driven
Use when a non-trivial decision sits under uncertainty and correctness matters more than speed. Not for patch review: use review. Not for plan attacks: use load-bearing-assumption-test.
pr-review-canvas-html
Use when asked to render a GitHub PR as a standalone review HTML page. Fetches PR data via gh API, renders diffs with move detection, and serves the artifact on a local port. Not for Cursor Canvas output — use pr-review-canvas.
resolve
Use when addressing review feedback in analyze or reception mode, or handling GitHub PR feedback in autonomous, interactive, or summary mode for a named PR target. Classifies, applies, replies, or summarizes comments.
property-test-authoring
Use when adding or improving property tests for invariants, oracles, parsers, algorithms, data structures, or smart-contract state machines. Encodes the strongest grounded property with domain-aware generators and pinned edge cases. Not for review — use property-test-review.