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 agentmods add skills/amajorai/spec.md/specnpx skills add amajorai/spec.md --skill specgit clone --depth 1 https://github.com/amajorai/spec.mdWhat 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 | $0.00050 | $0.03581 |
| Opus 5 | $0.00025 | $0.01791 |
| Sonnet 5 | $0.00010 | $0.00716 |
| Haiku 4.5 | $0.00005 | $0.00358 |
Grade B, and why
spec scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Enumerates other installed skillsmediumAgent snooping
Other skills' SKILL.md files reveal prompts, capabilities and secrets that should be invisible to peers.
ls ~/.claude/skills/party.md ~/.claude/skills/ship.md ~/.claude/skills/ship-fast.md 2>/dev/null How it starts
The opening of the file, as written. The whole thing — 358 lines — stays where its author put it; the contents beside it link to each section on GitHub.
👻 Spec
You are a spec architect. Your job is to take a task of any size or vagueness and break it into the smallest possible atomic units that a single agent can pick up and close independently. No overlap. No ambiguity. No gaps.
Task: {{args}}
Create tasks for every phase below with TaskCreate and TaskUpdate. Set addBlockedBy so each phase is blocked by the previous one. Mark each phase in_progress when you start it and completed when done.
Phase 0: Bootstrap
If {{args}} is empty (user ran /spec with no arguments): ask the user one question before doing anything else:
"What do you want to build or change? Give me a sentence or two and I'll take it from there."
Wait for the response. Treat it as {{args}} for all phases below.
Detect greenfield. Run via the Bash tool (Git Bash on Windows):
test -d .git && echo "HAS_GIT" || echo "NO_GIT"
ls -A 2>/dev/null | grep -vE '^(\.git|\.claude|README\.md|LICENSE|\.gitignore)$' | head -5
If no .git or the directory is essentially empty: set GREENFIELD=true. In greenfield mode:
- Skip codebase-exploration subagents in Phase 1+2. Go directly to the interview step.
- Require the user to confirm stack, framework, and deployment target as part of Phase 1+2 resolved areas.
- Make Unit 0 explicitly "Project scaffold: init repo, configure
<stack>, set up CI, push to GitHub." - After Phase 2.5: if no remote exists, offer to run
gh repo createbefore proceeding.
Phase 1 + 2: Explore-then-Interview Loop
These two phases run as a single loop. The goal is to arrive at a complete, unambiguous spec while asking the user as few questions as possible. Explore first, ask only what you cannot find.
Areas that must be resolved before proceeding:
- Scope: what is in and explicitly out of scope
- Target users and core use cases
- Existing codebase patterns, stack, and constraints (skip in greenfield; defer to interview)
- Acceptance criteria for the overall goal
- Atomicity threshold: roughly how large should a single unit be? Default is 200-600 LOC of net change across no more than 5 files, with 2-5 acceptance criteria. Confirm or adjust with the user.
- Existing GitHub issues or milestones to link against (to avoid duplicates)
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.
- 2d ago First seen · 358 lines · 50 tokens per session scan B f9e63fe5c273
spec is a skill published in the GitHub repository amajorai/spec.md (5 stars, last pushed 3mo ago), licensed MIT. It adds 50 tokens to every session and 3,581 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 1 finding (enumerates other installed skills). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
autoreview
Pre-commit/ship code review: Codex default; optional Claude or Pi.
babysit
Same-session monitoring loop for PRs, CI runs, tickets, and deployments using the monitorstart / monitorupdate / autonudgestop MCP tools. The loop re-injects your check instructions into THIS session on an idle interval — same context, same tools — and works from dashboard chat, Slack threads, and Discord DMs. Use…
convex-insights
Query a running Convex app's logs + health in natural language (official MCP): failures, slow/expensive functions, deploy causality — scoped, evidence-backed, with a dashboard deep link.
openclaw-design-system
Compatibility alias for existing OpenClaw installations that now applies Carapace semantic tokens, themes, shared CSS foundations, consumer adapters, and established local primitives.
happiness-skill
当用户问「怎么才能更幸福/为什么得到了还不满足/怎么减少焦虑」时调用。 核心理念: 幸福是缺憾感清空的默认状态, 是可训练的技能; 欲望是与自己的契约(得到前不快乐), 同时只留一个重大欲望; 活在当下。 不适用于: 临床抑郁等需要专业治疗的场景(本书方法不能替代医疗)。 Triggers: 幸福/不快乐/欲望/焦虑/知足/活在当下/happiness/desire/anxiety.
convex-add
Add a capability to the CURRENT Convex app — consults the served Convex capability catalog for always-current procedures (billing, crons, auth, agent, search, …); falls back to built-in hosting or @convex-dev component search. TRIGGER when the user runs /add, or asks to add hosting/publishing or any backend capability…