Borrowing it
Nothing to install: this file belongs to davidmatousek/tachi. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/davidmatousek/tachi/main/.claude/skills/~aod-deliver/SKILL.mdgit clone --depth 1 https://github.com/davidmatousek/tachiWrote 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/davidmatousek/tachi/aod-deliver)<a href="https://agentmods.dev/skills/davidmatousek/tachi/aod-deliver"><img src="https://agentmods.dev/badge/skills/davidmatousek/tachi/aod-deliver/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/davidmatousek/tachi/aod-deliver"><img src="https://agentmods.dev/badge/skills/davidmatousek/tachi/aod-deliver.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 177 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 | $0.00081 | $0.15396 |
| Opus 5 | $0.00041 | $0.07698 |
| Sonnet 5 | $0.00016 | $0.03079 |
| Haiku 4.5 | $0.00008 | $0.01540 |
Grade A, and why
~aod-deliver 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 6d 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:
- ~aod-deliver — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 1,184 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AOD Deliver Skill
Purpose
Close a completed feature with a structured retrospective that captures delivery metrics, surprises, and lessons learned. New ideas from the retrospective feed back into the Discover stage, completing the AOD Lifecycle loop.
Entry point: /aod.deliver
Prerequisites
- A feature branch exists with completed work (typically after
/aod.build [--no-security] [--no-simplify] [--no-docs]) .aod/spec.mdexists with the feature specification.aod/tasks.mdexists with task definitions.aod/scripts/bash/github-lifecycle.shis available for GitHub operations
Reference-Loading Contract
Several sections of this skill are stored in on-demand reference files under .claude/skills/~aod-deliver/references/. This mirrors the ~aod-run pattern.
Trigger idiom (used at every reference load site below):
MANDATORY: You MUST use the Read tool to load
references/X.mdbefore proceeding with [task]. Do NOT rely on memory of prior [X] content. If the file cannot be read, display an error and STOP.
Fail-closed rule: If a reference file cannot be read at a gate, display an error message and STOP execution. Do not attempt to continue from memory.
Re-read rule: If more than 50 lines of interstitial output have been generated since a reference was last read, re-read it before using its content.
Reference Decision Table
| Reference File | Holds | Load When |
|---|---|---|
references/deliver-flags.md |
Step 0 + Step 0g (flag parse + audit log) | Every run — at Step 0 |
references/delivery-lock.md |
Step 0.5 (acquire lock) + Step 14 (release lock) | Every run — at Step 0.5 and Step 14 |
references/close-and-document.md |
Step 12 (close issue) + Step 13 (prompt /aod.document) | Every run — at Step 12 |
references/render-tables.md |
Step 11c.1 Feature-139 render tables | Only when e2e_validation.status is set (E2E was entered) |
references/edge-cases.md |
Edge Cases, Common Rationalizations, Red Flags | Only when handling an edge case or checking a rationalization/red-flag |
What ships with it
5 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.
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.
- 6d ago First seen · 1,184 lines · 81 tokens per session scan A f084346e6884
~aod-deliver is a skill published in the GitHub repository davidmatousek/tachi (91 stars, last pushed 27d ago), licensed Apache-2.0. It adds 81 tokens to every session and 15,396 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 2026-09-03.
Other skills, from other repositories
usap-program-manager
Security program manager and passive lifecycle orchestrator (cs-security-program-manager). Use for security roadmap planning, program health scanning, debt tracking, facilitated reviews, or when the user says "security roadmap", "program planning", "security debt", "proactive scan", or "program review".
findings-tracker
../../../governance/findings-tracker/SKILL.md.
incident-commander
../../../response/incident-commander/SKILL.md.
contact-template
Contact / communications plan generator — primary operator, escalation chain, abort signal recipient, external SOC endpoint, blackout windows.
usap-ciso
CISO and executive security advisor (cs-ciso-advisor). Use for board reports, executive risk briefings, regulatory gap assessments, cyber insurance inputs, or when the user says "board report", "CISO brief", "executive summary", "risk posture for leadership", or "regulatory update".
usap-incident-responder
Active incident command and response (cs-incident-responder). Use when there is a confirmed or suspected active incident, for containment decisions, forensic investigation, SEV declaration, or when the user says "declare incident", "start IR", "we have an active breach", or "forensics".