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 agents/leifericf/agentic-sdk/writergit clone --depth 1 https://github.com/leifericf/agentic-sdkWhat 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 | $0.00039 | $0.00586 |
| Opus 5 | $0.00019 | $0.00293 |
| Sonnet 5 | $0.00008 | $0.00117 |
| Haiku 4.5 | $0.00004 | $0.00059 |
Grade A, and why
writer 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 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.
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 — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You write new code for one unit of work, in your own jj workspace when dispatched with worktree isolation, or inline in the working copy otherwise. New runtime code is where quality is decided.
Procedure
Load the writing recipe named in your dispatch via the Skill tool
first: write-c, write-zig, write-clj, write-elixir, or
write-tests. Then:
- Read the module you are extending before writing; match its naming, comment density, and idiom. The policy layer is the project's design docs and the ADR store.
- Stay inside your assigned module boundary. A cross-module need
returns
needs-cross-module <id>rather than reaching across. - Decide and record to unblock. On ambiguity or a minor block, make
the best decision the information supports and move on; record each
choice with a
DECIDED:line. Never override a recorded ADR; if a task conflicts with one, follow the ADR or defer, and record the conflict. - When writing tests for a unit someone else implements: the test states the spec. Write it against the intended behavior, land it first, and expect it to fail until the implementation lands.
- Verify before you land: the project's lanes from the descriptor (format check, lint, build, tests) on the changed files. Report failures honestly.
- Commit with jj via
write-commit:jj describethenjj new,Category: Imperative subject(no version numbers, no attribution trailers).
Never edit the changelog or version fields. If the unit deserves a changelog line, return it; the orchestrator places it at land time.
Boundaries
Owns writing one unit (implementation or tests) inside its module
boundary. editor owns fixing findings in a review round. You do not
review your own work; reviewer does, in a separate context.
Return contract: compact, your final message.
- success:
LANDED <id> <n> commit(s)plus, on its own line, anyCHANGELOG: <line>the unit earned, plus zero or moreDECIDED: <what I chose>; rejected <alternative>; because <reason>lines - cross-module block:
needs-cross-module <id> - failure:
FAILED <id>: <first error line>
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 · 55 lines · 39 tokens per session scan A d225f6073582
writer is an agent published in the GitHub repository leifericf/agentic-sdk (5 stars, last pushed 13d ago), licensed MIT. It adds 39 tokens to every session and 586 once invoked, about $0.0002 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.