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 HigorAlves/orc --skill doc-writinggit clone --depth 1 https://github.com/HigorAlves/orcWrote 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/higoralves/orc/doc-writing)<a href="https://agentmods.dev/skills/higoralves/orc/doc-writing"><img src="https://agentmods.dev/badge/skills/higoralves/orc/doc-writing/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/higoralves/orc/doc-writing"><img src="https://agentmods.dev/badge/skills/higoralves/orc/doc-writing.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 | $0.00058 | $0.02215 |
| Opus 5 | $0.00029 | $0.01107 |
| Sonnet 5 | $0.00012 | $0.00443 |
| Haiku 4.5 | $0.00006 | $0.00221 |
Grade A, and why
doc-writing 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 9d 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 — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Doc Writing — shared scaffolding
This skill holds the material common to every numbered design/spec document orc authors. The five type-specific skills — orc:prd-writing, orc:trd-writing, orc:rfc-writing, orc:adr-writing, orc:postmortem-writing — defer here for the shared parts and add only their own template, fields, and tone.
Read this first when authoring any of those docs. It answers: which doc type is this?, how is it laid out?, where does it get published?, and what are the review gates?
The doc-type map — which one is this?
These five document types form a lineage, each answering a different question at a different moment. Picking the wrong one wastes the reader's time and yours. Use this matrix to choose.
| Document | Primary question it answers | Audience | Stable / lives until |
|---|---|---|---|
| PRD | "What are we building, for whom, and why?" | PM, design, engineering, leadership | Product intent locked in |
| TRD | "What's the technical contract — interfaces, data, failure modes?" | Engineering | Engineering agrees on the contract |
| RFC | "How should we build it? Which alternative wins?" | Engineering (sometimes wider review) | Decision made (Approved / Rejected) |
| ADR | "We decided X; here's why." | Future-you and successors | Forever (until superseded) |
| Postmortem | "What did this incident teach us about the system?" | Engineering, on-call, leadership | Final once all P0 action items close |
| Plan | "In what order do we ship the slices?" | Whoever is implementing | Work shipped |
A typical lineage: PRD → TRD → (RFC if alternatives exist) → plan → ADRs as decisions lock in. A postmortem is off to the side — it's triggered by a production incident, not by the build pipeline, and it often spawns ADRs and doc updates downstream.
How each type is distinct
- PRD vs TRD — a PRD answers what & why for the user; a TRD answers what we'll build, technically (the interfaces, data shapes, failure modes). A PRD heavy with API shapes and schemas is a TRD mislabeled. A TRD with no PRD behind it is answering a question nobody asked.
- TRD vs RFC — a TRD commits to one contract; an RFC debates alternatives. Argument belongs in the RFC; the TRD records the contract that survived. A TRD that debates
pgbouncer vs. pgcatis really an RFC. - RFC vs ADR — an RFC is exploratory and opens debate before a decision; an ADR is the durable, post-hoc record of a decision already made. An approved RFC typically spawns 1–N ADRs (one per durable decision it locks in) plus one plan.
- RFC/ADR vs plan — RFC and ADR are about the design and why; a plan is the ordered work breakdown once design is settled. Don't write an RFC for something with clear precedent you'd implement in a day — write a plan or just do it.
- Postmortem vs systematic-debugging —
systematic-debuggingfixes the bug; the postmortem learns from the system around the bug (the gaps in tests, review, alerting, and process that let the bug reach production). The fix lands in code; the postmortem lands in the team's memory. - prd-writing vs orc:to-prd —
prd-writingauthors a PRD from scratch, interview-driven, and publishes todocs/prds/.orc:to-prdsynthesizes an existing conversation into a PRD shape and ships it to an issue tracker. "I have an idea, help me PRD it" →prd-writing. "We just talked this through; write it up" →orc:to-prd.
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.
- 9d ago First seen · 94 lines · 58 tokens per session scan A 9bbe8ea5cffe
doc-writing is a skill published in the GitHub repository HigorAlves/orc (6 stars, last pushed 12d ago), licensed MIT. It adds 58 tokens to every session and 2,215 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
taiyi-ui-design
A design-planning guide for describing how an application's user interface should look and behave. It produces a UI-DESIGN.md document covering layouts, components, interactions, accessibility, and error states.
taiyi-evolve
A workflow skill that compares the implemented code with the frozen design after development and testing. It records architecture changes and proposes updates to DESIGN.md, the document describing the intended system structure.
taiyi-diagram-c4
A code-scanning tool that builds C4 architecture documents from a repository. It separates facts observed in the code from conclusions inferred about the design and uses Mermaid diagrams as the source format.
flow-next-resolve-pr
Resolve PR review feedback. Fetches unresolved threads, triages, fixes, replies and resolves via GraphQL. Use when asked to address review comments.
flow-next-tracker-sync
Project a flow-next spec to a tracker issue (Linear, GitHub, GitLab, Jira) and reconcile two-way. Use when asked to sync to a tracker. NOT plan-sync.
writing-style
Use for technical communication - GitHub/GitLab tickets, PR/MR descriptions, issue comments, code review comments, commit messages. Direct, brief style with no AI-speak. NOT for README.md, public docs, or blog posts.