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 OutlineDriven/outline-driven-development --skill cli-for-agentsgit clone --depth 1 https://github.com/OutlineDriven/outline-driven-developmentWrote 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/outlinedriven/outline-driven-development/cli-for-agents)<a href="https://agentmods.dev/skills/outlinedriven/outline-driven-development/cli-for-agents"><img src="https://agentmods.dev/badge/skills/outlinedriven/outline-driven-development/cli-for-agents.svg" alt="Measured on agentmods" 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 | $0.00044 | $0.00906 |
| Opus 5 | $0.00022 | $0.00453 |
| Sonnet 5 | $0.00009 | $0.00181 |
| Haiku 4.5 | $0.00004 | $0.00091 |
Grade A, and why
cli-for-agents 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 today.
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:
- cli-for-agents — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 41 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLI for agents
Contract
| Field | Bound contract |
|---|---|
| Trigger | Human asks to build or review a CLI intended for coding agents |
| Authority | Read-only. No file, VCS, credential, paid, published, deployed, or remote mutation. Advice or review only. |
| Side effect | Chat output only: design advice or review findings. |
| Done | A report returned that prescribes flag-driven, pipeline-safe, idempotent CLI design. |
Inputs
- A CLI to review, or a build request describing the CLI's purpose and target agent runtime. Either is sufficient; both may be supplied.
- Optional: existing command surface, flag set, exit-code map, or invocation examples. When absent, state the assumption and review against the design rules below.
Procedure
- Bound scope: confirm whether this is a build (propose a design) or a review (judge an existing CLI). Do not mutate the CLI or any file; produce advice only. Done when: build or review intent is confirmed and stated in the report.
- Enumerate the command surface the agent will invoke: subcommands, flags, positional args, stdin, stdout, stderr, and exit codes. Done when: the full surface is listed or the missing surface is named.
- Check flag-driven control: every behavioral switch is a flag or subcommand, not an interactive prompt that blocks on a TTY. Flag names are stable, long-form, and discoverable via
--help. Reject hidden modes toggled only by interactive input. Done when: every switch is a flag or subcommand, or each violation is recorded. - Check pipeline safety: output meant for machine consumption goes to stdout as plain, parseable text; human diagnostics go to stderr. The CLI never hangs waiting on a TTY when stdin is not a terminal. Exit codes are explicit and documented: 0 success, non-zero for distinct failure classes. Done when: stdout/stderr split and exit-code map are verified, or violations are recorded.
- Check idempotence: re-running the same invocation with the same inputs produces the same result and side effects, or fails loudly with a non-zero exit when the operation is not safely repeatable. Done when: each state-changing command declares idempotence or names what makes a repeat unsafe.
- Check agent ergonomics: deterministic output ordering, no color or progress decoration on stdout unless explicitly flagged on, machine-readable
--jsonor equivalent where the agent parses output, and a single canonical invocation form per action. Done when: all four ergonomics rules pass or each violation is recorded. - For a build request, propose the command surface, flag set, exit-code map, and stdout/stderr contract that satisfy rules 3-6. For a review, record each violation against the rule it breaks and prescribe the fix. Done when: the proposed surface or the per-violation fix list is complete.
- Stop at the report. Do not edit the CLI, run it, or widen scope into implementation, testing, or packaging. Done when: the report is returned and no file was mutated.
What ships with it
1 file 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.
- today Changed dd3499cb4364
- 2d ago First seen · 41 lines · 44 tokens per session scan A 6f9ac5b69a43
cli-for-agents is a skill published in the GitHub repository OutlineDriven/outline-driven-development (52 stars, last pushed yesterday), licensed Apache-2.0. It adds 44 tokens to every session and 906 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 2026-09-03.
Other skills, from other repositories
slicing-code-context
Use when an exact symbol, path, entrypoint, or line range can bound a focused code question or patch proposal under a fixed source budget. Not for source changes or broad repository exploration.
askme
Explore intent with Verbalized Sampling before planning. Use when the task is ambiguous, you need maximum clarifying questions, or you want collaborative or adversarial mode via /askme.
handoff-prompt
Use when the user asks for a handoff, delegation, or clipboard-ready prompt for another agent: a standalone path-free prompt copied to the clipboard, confirmed by title. Not for session-snapshot briefs — use handoff; never remote, credential, publish, deploy, or irreversible.
autopilot
Run a hands-off plan-to-ship pipeline by chaining existing skills. Use when the user says "autopilot", "take this from plan to shipped", "run the whole pipeline", "hands-off ship it", or "do the end-to-end build".
llm-self-loop
Use when a button click, dashboard check, or human verdict sits inside an iteration loop. Replaces it with an autonomous gate or moves non-automatable work outside the loop. Don't use for remote, credential, publish, deploy, or irreversible changes.
plan-review-tune
Use when plan-mode enforcement hooks intercept a plan review or the user runs /plan-review-tune to tune which plan-review questions fire. Persists the tuned question flow to a local question-registry and hook configuration. Not for auditing a plan against code — use plan-review.