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 makifbaysal/tasktrooper --skill spec-authoringgit clone --depth 1 https://github.com/makifbaysal/tasktrooperWrote 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/makifbaysal/tasktrooper/spec-authoring)<a href="https://agentmods.dev/skills/makifbaysal/tasktrooper/spec-authoring"><img src="https://agentmods.dev/badge/skills/makifbaysal/tasktrooper/spec-authoring/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/makifbaysal/tasktrooper/spec-authoring"><img src="https://agentmods.dev/badge/skills/makifbaysal/tasktrooper/spec-authoring.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00038 | $0.01476 |
| Opus 5.5 | $0.00015 | $0.00590 |
| Sonnet 5.5 | $0.00008 | $0.00295 |
| Haiku 4.5 | $0.00004 | $0.00148 |
Grade A, and why
spec-authoring 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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Spec Authoring
Overview
The spec is not a document of its own any more: it is the context and design sections of the ONE analysis report the analiz task produces (analiz-html-report) — an HTML document attached with add_task_document, format: "html", titled analiz: <YYYY-MM-DD> <topic>. Never attach a separate spec: … document, never write it to a file and never commit it; shelling out to echo/heredoc to fake a file is how a run gets killed for repeating itself. The spec is the contract the plan section and every implementation task will be checked against.
These sections must name real files, symbols, and interfaces you found with codebase_search / grep_code / get_repo_tree / expand_symbol_context. A run that attaches an analiz document without a single successful exploration call is rejected by the system, not by a reviewer.
Structure
Where each part lives in the report: Context / Goal feeds summary and context; Architecture through Out of scope are subsections of design (each an <h3> with its own id). Scale each part to its complexity — a few sentences if straightforward, up to 200–300 words if nuanced; Decision record and Security & data are each written only when they apply (see their conditions below):
- Context / Goal — the problem, who has it, what outcome closes it.
- Architecture — the chosen approach, and in one line why it beat the alternatives.
- Components — each unit with clear boundaries: what it does, its interface (exact names and types), what it depends on.
- Data flow — how a request/value moves through the components, including where validation happens.
- Error handling — what fails, how each failure surfaces, what the user sees.
- Testing approach — what proves each component works: unit, integration, end-to-end.
- Decision record — for a choice worth remembering (new dependency or framework, a datastore or schema shape, an API/contract pattern, auth or security architecture, a cross-repo integration — skip it for a bug fix or config change):
<h3 id="design-decision">with Problem (1–2 sentences) · Decision drivers (bullets) · Options (table: option | fits drivers | cost/risk) · "Chosen: X, because …" · Consequences (Good, because … / Bad, because …). If the repo keeps its own ADRs (docs/adr,docs/decisions,adr/), add a plan step in that repository that writesNNNN-<title>.mdin the repo's own format. - Security & data — who may call each new interface, which inputs are attacker-controlled, what data leaves the system (to a log, a third party, a client bundle), and what is logged (security-review has the full checklist for any new endpoint or data flow). Any schema or contract change also gets its expand/contract shape and consumer list here (migration-and-contract-review).
- Out of scope — explicitly named items deferred or excluded.
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.
- 4d ago Changed 04917c85b6fd
- 8d ago First seen · 62 lines · 38 tokens per session scan A b6f20c1f6a18
spec-authoring is a skill published in the GitHub repository makifbaysal/tasktrooper (113 stars, last pushed yesterday), licensed Apache-2.0. It adds 38 tokens to every session and 1,476 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 2026-10-05.
Other skills, from other repositories
github-multi-repo
Multi-repository coordination, synchronization, and architecture management with AI swarm orchestration.
architecture-design
Use this skill to design system architecture with documented trade-offs and decisions before any implementation begins. Activates when a feature, system, or component requires non-trivial structural decisions about how components interact, what data stores to use, or how to decompose a problem.
repository-discovery
Use this skill to systematically explore an unfamiliar codebase before making any changes. Activates when starting work in an unknown repository, adding a new feature, debugging without prior context, or when no CONTEXT.md exists. Produces a grounded understanding of architecture, conventions, and constraints.
rapid-prototyping
Use this skill to build disposable, rapid proof-of-concept prototypes and spikes to test feasibility, validate API designs, or explore library capabilities before committing to a full production architecture.
ramp
Fetch and follow Ramp's published agent playbooks inside Paperclip, with mandatory approval gates for spend, incorporation, cards, account setup, and other financial actions.
paperclip-create-agent
Create new agents in Paperclip with governance-aware hiring. Use when you need to inspect adapter configuration options, compare existing agent configs, draft a new agent prompt/config, and submit a hire request.